diff --git a/.agent-taskgraph/DECISIONS.md b/.agent-taskgraph/DECISIONS.md index c0bd146c..9df58cb3 100644 --- a/.agent-taskgraph/DECISIONS.md +++ b/.agent-taskgraph/DECISIONS.md @@ -1,164 +1,12 @@ -# Decisions +# 决策记录 -## 2026-08-10 Product Direction +## 2026-08-13 / P9.1 持续开发批次 -- Original product: `Aetherbound Guild / 裂脉远征团`. -- Familiar, readable magical guild and job fantasy; no Dragon Quest or reference-game expression is copied. -- Complete scope: 30 named characters, 12 base jobs, 24 advanced jobs, 100 enemy identities, 8 chapters and 320 equipment/items. -- Target playtime: 25-35 hour first campaign, 60-80 hour build completion and roughly 100 hour full collection/high difficulty. -- Landscape 16:9; full 2D Japanese hand-drawn direction using layered rigs, equipment layers, authored key poses and unique portraits/illustrations. -- Formal implementation begins only after the complete design package, HTML gallery and interactive design prototype pass Owner review. - -## 2026-08-10 Execution Decision - -- Use Agent TaskGraph protocol 0.8.0-beta.10. -- Run three parallel non-overlapping design workers, then presentation, integration and an independent reviewer. -- All worker and reviewer sessions use Codex `gpt-5.6-sol`, `xhigh`, project-level `yolo`. -- Native `wait_agent` is the monitoring mechanism; no shell sleep or passive chat waiting. - -## 2026-08-10 R2 Runtime Recovery - -- R1 correctly stopped on a PMO worktree-path mismatch. Goal paths were corrected, context revisions advanced to 2, and dispatch IDs advanced to r2. -- Because the original collaboration follow-up did not yield a valid recoverable ACK, the same local Codex runtime was relaunched through `codex exec` with unchanged `gpt-5.6-sol`, `xhigh`, `yolo` and isolated worktrees. -- CLI session/JSONL IDs are the r2 identity evidence. R1 draft output is not accepted unless the r2 worker inspects, completes, commits and submits it. -# ABG-SYSTEM-REPAIR - -| 时间 | 决策 | 决策者 | 影响 | -|---|---|---|---| -| 2026-08-10 | 接受 HAPI reviewer 对 `22b21f8` 的 FAIL，批准同一 systems-designer 角色进行一次有界前向修复；为保持活跃 batch 的已确认 runtime 与 validator 一致，worker 复用本批 Codex native profile，修复后的独立 reviewer 仍为 HAPI 可见；固定 `gpt-5.6-sol / xhigh / yolo` | Owner：`继续` | graph-r2 新增 `system_design_repair`；Frozen 产品范围不变；表现设计依赖改为修复节点 | - -# Stop External Orchestration - -| 时间 | 决策 | 决策者 | 影响 | -|---|---|---|---| -| 2026-08-10 | 停止本批 AgentGraph、所有外部 worker/reviewer 和后台监控，后续在当前 Owner 会话直接整合与修复 | Owner：`停止了吧，就在当前会话进行吧` | Batch 01 关闭；候选提交保留；无会话继续执行；不再使用 queue 状态驱动本轮设计工作 | - -# Recruit-Line Redesign Batch 02 - -| 时间 | 决策 | 决策者 | 影响 | -|---|---|---|---| -| 2026-08-10 | 以参考游戏已验证的横向队列、持续招募、装备构筑和自动战斗为玩法骨架；删除固定 30 名剧情角色、关系系统、固定六人、三路线和战中 Directive | Owner | 旧产品合同作废，必须前向重写 | -| 2026-08-10 | 完整原创规模采用 12 基础 + 24 进阶 + 6 隐藏职业、36 特性、320 装备、60 神器、100 普通/精英敌人、16 Boss、8 区域 | Owner：`对可以这样设计` | design-spec-r2 冻结 | -| 2026-08-10 | 允许开启多个可见会话；两个无冲突写入 worker 并行，Owner 手动开启的 `审计 Aetherbound 复刻差距` 会话保留为后置只读 reviewer | Owner | HAPI visible Codex gpt-5.6-sol/xhigh/yolo；UI 工作等待系统与内容验收 | - -# Recruit-Line Redesign Batch 02 Parallel Expansion - -| 时间 | 决策 | 决策者 | 影响 | -|---|---|---|---| -| 2026-08-10 | 本批可在真实职责和无写入冲突前提下扩展到 10 个以内可见会话；先增设 UX、美术、音频与设计验证四条设计线 | Owner：`可以再多几个会话，10个以内我觉得都OK` | design-spec-r3 / graph-r4；六个 worker 并行，一个独立 reviewer 后置；HTML 与运行时仍冻结 | - -# Recruit-Line Redesign Batch 02 Dual Review - -| 时间 | 决策 | 决策者 | 影响 | -|---|---|---|---| -| 2026-08-11 | 六条设计线完成后，不再新增重复写作者；新增系统完整性与体验生产两个并行只读 reviewer，PMO 汇总后进入 Owner Gate | Owner：`可以再多几个会话，10个以内我觉得都OK` | Team rev-4 / graph-r5；项目相关可见会话总量不超过 10；HTML、引擎、最终媒体、打包与发布继续冻结 | - -# Recruit-Line Redesign R2 Review Routing - -| 时间 | 决策 | 决策者 | 影响 | -|---|---|---|---| -| 2026-08-11 | 接受 Systems R2 对 `7054fe6` 的 Gate 5 FAIL；保留既有“无隐藏 miss chance”，先定义确定性 action outcome，再并行同步 content/art/audio，不新增 accuracy/evasion 概率系统 | PMO，依据 Frozen 可理解性与既有战斗权威 | graph-r6；复用原 worker 角色；新固定 revision 后必须使用两个 fresh reviewer | -| 2026-08-11 | Experience R2 因 context 引用了不存在的 `REFERENCE_ANALYSIS_AND_LESSONS.md` 判为 INVALID；不作产品判断、不复用会话 | PMO reviewer protocol | 修正未来 R3 context 为现存 `REFERENCE_RUNTIME_CENSUS.md` 与 `TRANSFERABLE_STRUCTURE_LESSONS.md` | -# 2026-08-11 - Experience R3 Repair Routing - -- Decision: accept `ABG-V2-EXPERIENCE-REVIEW-R3 FAIL` as a product-design failure on immutable `d454120`. -- Reason: run creation has no provisional choose-two presentation state, party-line direction conflicts, recruit content/art schemas lack a crosswalk, audio omits `OUT-006` and miscounts screens, and the lessons authority retains fixed-cast scope. -- Route: reuse the persistent UX, Art, Audio and Validator sessions with disjoint writes; add one visible Reference repair session. Systems R3 continues independently on the immutable source. -- Frozen: no HTML, runtime, final media, packaging or release. Repaired documents require a new immutable revision and fresh experience R4 review. - -# 2026-08-11 - Systems R3 Gate 6 Repair Routing - -- Decision: accept the immutable `d454120` Systems R3 FAIL and route only the missing fixed-seed anti-dominance proof to the existing Systems role. -- Reason: battle, transaction, formula, pacing and solvency checks passed; the earliest failure is absent executable seed fixtures and policy outcomes. -- Route: one new fixture JSON, one simulation tool and the economy authority only; later integration creates fresh Systems R4 and Experience R4 reviews. -- Frozen: no stochastic accuracy/evasion, unrelated rebalance, content expansion, HTML, runtime, media, packaging or release. - -# 2026-08-11 - R3 Repair Runtime Reconciliation - -- Art candidate `7f8d4b4` and Audio candidate `9f63e46` were edited, committed and pushed before PMO observed their exact standalone rev-5 identity ACKs. Their later ACKs reconcile identity but do not erase the process-order deviation or independently accept the product changes. -- Reference candidate `565ba2e` reached its legal terminal with clean/upstream evidence. Systems HAPI `22caeb51-307e-41a6-af3e-875b8e37ba6b` returned the exact ACK before editing and is active. -- Current TaskGraph state validation now accepts the six current tasks' dispatch/runtime/ACK fields. Remaining errors are pre-existing review-role concurrency/history debt; the installed graph validator also rejects the repository's already-approved `fork` node kind. Do not expand this bounded repair into a wholesale legacy-ledger or graph-schema migration. -- UX `388d90b`, Art `7f8d4b4`, Audio `9f63e46`, Reference `565ba2e`, and Validator `ec5e234` passed PMO write-scope/evidence inspection and were integrated without conflict as `14ea46e`, `03547a7`, `7071887`, `96ea49a`, and `8a6f0c8`. This is integration, not R4 or Owner acceptance. -- Systems `cb53794` was integrated as `b65646c`; anti-dominance self-test and canonical replay pass. Combined validation then exposed three validator-only false negatives against real repaired prose/tables while its generated fixture remained green. R4 is blocked on a two-file H1 parser/regression repair; product authorities remain frozen. - -# 2026-08-11 - H1 Integration Acceptance And R4 Freeze - -- H1 candidate `49e128f` changes only the validation contract and executable; PMO integrated it as `bcb2931` and independently reran the combined validator and anti-dominance gates. -- Text and JSON validation return zero diagnostics, the validator self-test remains 20 cases, the five contradiction mutations remain rejected, and the five fixed policy seeds preserve a non-dominant winner spread. Result SHA-256: `17154371d4b93035296d1219f012f381af2e10a5f2c1b3f66bcfcd7ad4d85352`. -- Create one PMO documentation closure and use that exact pushed revision as the sole immutable source for fresh Systems R4 and Experience R4 reviewers. No R1-R3 partial judgment may be reused as PASS. -- HTML, engine work, final media, packaging and release remain frozen. Dual reviewer PASS leads only to PMO synthesis and the Owner design gate. - -# 2026-08-11 - R4 Review Failure Routing - -- Accept Systems R4 FAIL on immutable `db932cf`: provisional generated Recruits are selected before the only declared run seed exists, so offer reproduction, selection receipts and crash retry are undefined. -- Accept Experience R4 FAIL on the same immutable source: `PF-H01..H06` content mechanics/production briefs and art parent/silhouette/motion bindings are inconsistent, with unresolved-binding language still present. -- Route two disjoint minimal repairs in parallel: seed/transaction authority across the exact product/content documents, and hidden-profession binding inside the art catalog only. Do not broaden balance, content count, UI, audio, HTML, runtime or media scope. -- After integration, add deterministic validator negatives for both R4 contradiction classes and create a new fixed revision. Only fresh dual reviewer PASS may reach the Owner gate. - -# 2026-08-11 - R5 Repair Attempt-1 Identity Invalidation - -- Both attempt-1 repair sessions edited before returning the exact standalone revision-bound `IDENTITY_READY`; PMO STOP messages did not prevent later work. Both sessions are archived and their runtime evidence is retained as `runtime-evidence-attempt1-invalid.json`. -- The run-creation attempt-1 dirty worktree is preserved unchanged as invalid history. Hidden-art commit `c558d19c1c96a07ab9e7caa23f495875126bc113` is explicitly forbidden from integration. -- One final attempt is authorized under the existing failure route: fresh clean `-a2` worktrees at `0b0f6ad`, context revision 2, dispatch attempt 2, exact identity ACK before any edit. A second identity-process failure stops the branch and requires Owner action. - -# 2026-08-11 - R4 Repair Integration And H2 Route - -- Attempt-2 workers returned exact identity ACKs before any edit, stayed within the three-file transaction scope and one-file hidden-art scope, pushed clean candidates `ad0a57d` and `fef9057`, and were archived after their legal terminals. -- PMO integrated them as `bba0c43` and `8102bd2`. This closes the two bounded repair nodes only; it does not grant complete design or Owner acceptance. -- Continue to the approved `v2_r4_validator_h2` node, then freeze one new source for fresh Systems R5 and Experience R5 reviews. HTML, engine, media production, packaging and release remain frozen. - -# 2026-08-11 - Validator H2 Staffing - -- Add one fresh task-scoped `design-validator-r5-h2` role under Team rev-8 rather than reviving the retired H1 validator identity. -- H2 owns only the validation contract and executable. It must preserve the 20 existing self-test cases and add exactly four isolated R4 contradiction cases before any R5 reviewer is created. - -# 2026-08-11 - Validator H2 Attempt-1 Invalid - -- HAPI `0ab9a014` passed runtime verification but did not emit the required exact standalone identity ACK as its first agent response. PMO observed a still-clean worktree, sent STOP, archived the session and classified the attempt `INVALID_IDENTITY_PROCESS`. -- No H2 edit, test, commit, push or judgment exists. R4 product repairs remain integrated at `8102bd2`; R5 dual review remains blocked on H2. -- The graph/Goal limit is 1/1. A fresh attempt requires Owner authorization to advance the dispatch/context and max-attempt boundary. - -# 2026-08-11 - Validator H2 Attempt-2 Recovery Authorized - -- Owner explicitly answered `OK的啊继续吧`, authorizing the H2 attempt boundary to advance from `1/1` to `2/2` and a fresh replacement session. -- Retain attempt-1 files and runtime evidence as immutable INVALID identity-process history. The retired `design-validator-r5-h2` role and archived HAPI `0ab9a014` must not be reused. -- Add task-scoped replacement role `design-validator-r5-h2-a2` under Team `rev-9`, a new worktree/branch, context revision 2 and dispatch attempt 2. Exact standalone `IDENTITY_READY` remains a hard pre-edit gate. -- Product scope and acceptance do not change: only the validation contract and executable may change, exactly 24 self-tests are required, and fresh R5 dual review remains downstream. HTML, engine, media, packaging and release stay frozen. - -# 2026-08-11 - Validator H2 Attempt-2 Identity Invalid And Exhausted - -- HAPI `2cf38e48-f8f9-4c61-adff-b822831c17c7` passed pre-dispatch runtime verification and received the generated rev-9 bootstrap. Its first agent response was an analysis sentence rather than the exact standalone `IDENTITY_READY`. -- The real 90-second HAPI timer/observe cycle (`functions.wait` cell 1986) exposed the mismatch. PMO sent STOP, confirmed the fresh worktree remained clean at `b0e3475`, and archived the session before any validator edit, test, commit or push. -- Classify attempt-2 as `INVALID_IDENTITY_PROCESS`; it makes no product or validator judgment. The H2 `2/2` limit is exhausted, so R5 reviewers remain uncreated and the queue stops for a new Owner decision. - -# 2026-08-11 - Pre-dispatch Recovery Does Not Consume Product Attempts - -- Owner directed PMO to continue without repeatedly requesting authorization for identical process recovery. -- Correct the earlier accounting: a verified HAPI session that fails identity before ACK, stays outside `active`, and leaves the Goal worktree clean with zero edits/tests/commits does not consume `graph.max_attempts`. Both prior H2 sessions are retained as INVALID process history; H2 product execution remains `0/2`. -- PMO may automatically create another fresh session when runtime, machine, model, effort, permission, visibility, write scope, cost and Human Gates are unchanged. Any product edit, scope/cost/permission expansion, paid action, packaging or release boundary still requires the normal gate. -- Recovery 3 uses a two-stage dispatch: first message carries only Role/Team/Goal/Context/continuity identity and requests the exact ACK; after ACK is observed and recorded, PMO sends the rendered full Goal bootstrap. - -# 2026-08-11 - Validator H2 Integrated And R5 Review Unblocked - -- Recovery-3 HAPI `1546960c-baf0-4a36-bd71-b1c39bc577b7` returned the exact rev-10 ACK before any read/edit, then completed the bounded two-file Goal. -- Candidate `fb20b663df97bb685887e5bef9bb6a3256fa7580` changes only the validation contract and executable. It reports `cases=24`, integrated `errors=0`, byte-identical JSON SHA-256 `4feb0666acc639a163fefadda40a4cda3a6e5caf0db58fa01432c773ec8d6e38`, five anti-dominance seeds and all four isolated R4 diagnostics. -- PMO confirmed exact scope, preserved prior gates, integrated the candidate as `7dc584d62fc4323746c927f5dc370e77f992f0c7`, pushed clean/upstream and archived the worker. -- Freeze one new documentation closure as the shared immutable R5 source. Create fresh Systems R5 and Experience R5 read-only reviewers; neither may inherit R4 PASS/FAIL as current judgment. - -# 2026-08-11 - R5 Dual Independent Review Staffing - -- Use closure `69f6715b6f47b507fb07436ee1edf77df7e5adab` as the one immutable review source containing integrated H2 `7dc584d` and its PMO closure. -- Add fresh task-scoped Systems R5 and Experience R5 reviewer roles under Team rev-11. Both use visible HAPI Codex `gpt-5.6-sol / xhigh / yolo`, separate worktrees and zero repository writes. -- Systems reruns all gates from zero with the repaired seed/draft/promotion transaction. Experience reruns all gates from zero with exact hidden-profession production bindings. R4 findings are history, not current PASS. -- A PASS from one reviewer cannot compensate for FAIL/INVALID from the other. Only dual PASS proceeds to PMO synthesis and Owner design review. - -# 2026-08-11 - Experience R5 Environment Recovery - -- Experience R5 attempt-1 used zsh special variable `path` in its first archive-verification harness and failed before source judgment, then continued after PMO STOP. Classify the whole attempt INVALID and archive HAPI `c52000b5-2021-48cb-8d68-299f36876b8c`. -- Its assigned worktree remained clean at `e4ef2b0`; no product judgment or repository write is reusable. Automatically create one fresh replacement reviewer on the same immutable source and unchanged read-only Goal. - -# 2026-08-11 - Final Design Package Closure - -- Owner explicitly prioritizes fast completion of the complete pre-development product package and final reviewable output, not engine implementation. -- Preserve Experience R5 Gates 1-7 PASS and repair only its earliest real blocker: exact `RG-01..08` to `ART-REG-01..08` production mapping plus validator regression coverage. -- In parallel, replace the superseded 82-screen coverage audit with the current-authority 50-state landscape HTML prototype. No engine, bulk final media, packaging or release work is authorized. +- Owner 要求继续开发，并使用 Agent TaskGraph 并行组织游戏策划、技术架构、设计师和多个开发。 +- 采用分波次并行，不允许多人同时修改 `runtime/main.gd`。 +- 第一波只产出游戏策划、UX/视觉、技术架构三份互斥规格。 +- 第二波在规格汇合后先拆运行时基础，再并行开发 domain、UI 和 tests。 +- 旧 `recruit-line-redesign-02` 整体归档，避免过期 active 状态和废弃图节点污染 P9.1。 +- 继续使用 visible HAPI Codex `gpt-5.6-sol / xhigh / yolo`；无静默 fallback。 +- HAPI 控制面超时不得伪装成已派发。任务留在 inbox，恢复后生成 fresh evidence 和 ACK。 +- 每个完成的玩法模块仍要有新 H5；本批最终 H5 必须在独立 reviewer PASS 后发布。 diff --git a/.agent-taskgraph/PROJECT.md b/.agent-taskgraph/PROJECT.md index 2f2867e5..51d4ef9a 100644 --- a/.agent-taskgraph/PROJECT.md +++ b/.agent-taskgraph/PROJECT.md @@ -2,84 +2,86 @@ ## 是什么 -`Aetherbound Guild / 裂脉远征团` 是原创横屏 2D 日系手绘魔法公会远征策略游戏。当前只允许完整开发前设计和 HTML 原型；正式引擎开发冻结。 +`Aetherbound Guild / 裂脉远征团` 是原创横屏 2D 日系手绘魔法公会自动战斗游戏。 +Phase 8 产品设计已建立；Phase 9 正在用 Godot 逐模块实现。P9.0R10 已证明任意两名 +初始 Recruit 都能完成前两战，P9.1 要把六种组合接入同一条完整首局，并拆分单体运行时。 -## 技术栈与目录结构 +## 当前技术栈与结构 -当前无运行时引擎。Markdown/JSON/CSV 是设计权威，HTML/CSS/JS 只用于设计原型。`docs/reference`、`docs/product`、`docs/content`、`docs/presentation`、`docs/prototype`、`docs/reviews` 按职责隔离。 +- Godot 4.6.1，横屏移动端，中文优先，真实运行时代码位于 `runtime/`。 +- `runtime/main.gd` 当前约 3700 行，是 P9.1 必须降低的协作和回归风险。 +- `docs/product/` 与 `docs/presentation/` 是完整设计权威；P9.1 只落地首局薄切片。 +- `prototype/` 是已接受的 HTML 设计参考，不是当前游戏实现。 +- `.agent-taskgraph/` 由 PMO 独占；worker 不修改队列、角色、状态和决策台账。 -## 硬性规范 +## 当前批次 -- 授权商业参考只提炼结构，不复制名称、文案、公式、布局、代码、图片或音频。 -- 完整设计包独立审查和 Owner 审核前，禁止运行时代码、引擎项目、批量最终资产、打包和发布。 -- PMO 独占 `.agent-taskgraph/` 与根共享文件；并行 worker 只写 Goal 指定目录。 -- 每条内容必须有机械作用、交互、进度位置、表现、资产/音频和验证字段；禁止 filler。 - -## 需求与执行授权策略 - -| 配置 | 项目选择 | +| 配置 | 已确认值 | |---|---| -| Agent TaskGraph 协议版本 | 0.8.0-beta.10（本活跃批次冻结；安装版 beta.12 只提供兼容 helper，不迁移协议） | -| Source baseline | READY: `/Users/wusumac/Documents/game/aetherbound_guild` at `ac0dec25848dbc6fe54f8de07c8396b01d8d3a89`, branch `main`, clean; redesign worktrees created only after the PMO baseline commit | -| 单任务快速路径 | 允许；本批不适用 | -| 复杂任务规格冻结 | `.agent-taskgraph/spec.md` design-spec-r3 已由 Owner 冻结 | -| 自主创建可见会话 | recruit-line-redesign-02 最多 10 个项目相关可见会话；完成六条设计线后新增两个独立 reviewer 已授权 | -| 派发预览授权 | 本批固定 HAPI 可见 Codex gpt-5.6-sol / xhigh / yolo 已授权 | -| 编制变更授权 | Owner 允许本批总会话数不超过 10；双 reviewer 分工已批准，仍须只读隔离、无写入冲突与真实 Goal | -| 默认 Human Gates | 依赖安装、删除、付费扩量、正式开发、打包、发布 | -| 已启用可选适配器 | HAPI：本批两个 worker 与 reviewer 均须用户可见、可接管 | -| Runtime 证据策略 | spawn 后验证线程、worktree、model、effort、permission 和 ACK；不匹配保持 inbox | -| 上下文模式 | balanced | -| Context 必读项上限 | 8 | +| 批次 | `abg-p9-1-continuous-development-01` | +| 协议版本 | `0.8.0-beta.12` | +| Source baseline | `f0b7d7b0234d8ad304f7b9d39fa99ded1dbfc27d`，`main`，clean/upstream | +| Frozen spec | `.agent-taskgraph/spec.md` revision `p9.1-spec-r1` | +| 当前阶段 | 先并行规格，汇合后再并行实现 | +| 最大同时工作会话 | 3 个 worker；PMO 独立监督 | +| Owner 授权 | 2026-08-13：使用 agent graph，并行安排策划、技术架构、设计师和多个开发 | +| Human Gates | 玩法理解、乐趣、视觉、听感、物理设备、打包、商店与发布 | -## Execution profile（派发硬门） +## Execution Profile | Execution profile | Confirmed value | |---|---| | Execution profile status | CONFIRMED | -| Execution profile confirmed by/at | Owner / 2026-08-10T22:38:28+08:00 / recruit-line-redesign-02 | -| Runtime choice confirmed by/at | Owner / 2026-08-10 / multiple visible HAPI sessions for the redesign batch | +| Execution profile confirmed by/at | Owner / 2026-08-13 / current P9.1 request | +| Runtime choice confirmed by/at | Owner / visible HAPI sessions requested for agent graph | | Execution runtime | hapi | -| Execution control | HAPI Hub helper spawn/inspect/message; visible sessions in Owner sidebar | -| Execution machine | local: id=a15888ad-d405-4884-a407-c28ddd8b397c; host=wusumacdeMac-mini.local | +| Execution control | HAPI Hub helper spawn/inspect/message; Owner sidebar visible and takeover-capable | +| Execution machine | local: id=`a15888ad-d405-4884-a407-c28ddd8b397c`; host=`wusumacdeMac-mini.local` | | Execution flavor | codex | | Model selection policy | fixed | -| Fixed model/effort | model=gpt-5.6-sol; effort=xhigh | -| Model catalog evidence | codex-cli 0.146.0; native catalog accepted gpt-5.6-sol with xhigh; per-session JSONL retained | +| Fixed model/effort | model=`gpt-5.6-sol`; effort=`xhigh` | | Execution permission | yolo | -| Permission scope | runtime-only; Human Gates remain | | Execution visibility | visible | -| Execution fallback | none; failed HAPI spawn remains inbox and is reported | -| Pre-dispatch recovery | Identity/runtime failure before ACK with clean worktree and zero Goal edits is PMO-owned recovery, not a product attempt; retry automatically while runtime/scope/model/permission/cost remain unchanged | -| Monitoring wait primitive | HAPI completion/blocker event + real timer cell; startup 90s, steady 600s; resume only with functions.wait(real cell_id) | -| Monitoring observe primitive | inspect_peer/session metadata/messages + exact task ledger reconciliation | -| Monitoring target evidence | Region-art R6 HAPI `c63c1b13-8f65-479c-ad65-b2c998713a2d` and final-prototype R6 HAPI `397e65db-795c-46c4-af7a-5eccf23ab56e` pre-dispatch VERIFIED | +| Execution fallback | none; HAPI control-plane failure leaves tasks in inbox | +| Pre-dispatch recovery | Clean, zero-edit runtime/identity failure may be retried without consuming a product attempt | +| Monitoring wait primitive | HAPI completion/blocker event + real timer cell; steady interval 600s | +| Monitoring observe primitive | `inspect_peer`/Hub metadata/messages + task ledger reconciliation | +| Monitoring target evidence | Current P9.1 HAPI Session IDs from active/review ledgers only | -## 共享文件锁 +Every new session requires fresh `runtime-evidence.json`, a delivered Role bootstrap and the exact +`IDENTITY_READY` before its task becomes active. A CLI binary or historical session is not proof. -PMO 独占 `.agent-taskgraph/`、root README/AGENTS and cross-domain indexes. `ABG-V2-CORE-SYSTEMS` exclusively writes four named files under `docs/product/`; `ABG-V2-CONTENT-MATRIX` exclusively writes `docs/content/`; UX, art and audio workers each own disjoint named files under `docs/presentation/`; the validator owns only `docs/reviews/DESIGN_VALIDATION_CONTRACT.md` and `tools/validate_predevelopment_design.py`. R3 repair reuses disjoint owners: UX writes the two UX authorities, art writes only the art catalog, audio writes only the audio catalog, reference writes only `TRANSFERABLE_STRUCTURE_LESSONS.md`, validator writes only its contract/tool, and Systems writes only `ECONOMY_AND_BALANCE.md` plus the new anti-dominance fixture/script. No worker edits `docs/prototype/` in this batch. +## 共享写入边界 + +- PMO only: `.agent-taskgraph/`, `CURRENT_CHECKPOINT.md`, P9.1 parent Goal/contract and acceptance records. +- Expert wave uses three separate Markdown outputs under `docs/runtime/p9_1/`. +- Foundation worker alone may extract shared APIs and change `runtime/main.gd` during extraction. +- Domain, UI and test workers run after foundation integration and own disjoint paths. +- Integration worker alone wires parallel products into shared entry and runner files. +- Reviewer is read-only. H5 publishing occurs only after reviewer PASS. ## 岗位工作区映射 -| 岗位 | 负责模块/目录 | 常用验收命令 | +| 岗位 | 负责范围 | 验收重点 | |---|---|---| -| reference-researcher | `docs/reference/` | evidence labels and local-link audit | -| systems-designer | `docs/product/` | contract consistency and decision audit | -| content-architect | `docs/content/` | exact-count/cross-reference/no-filler audit | -| presentation-designer | `docs/presentation/`, `docs/prototype/` | state/screen/asset coverage | -| independent-reviewer | `docs/reviews/` | fixed-revision read-only review | +| p91-game-designer | `docs/runtime/p9_1/GAMEPLAY_EXPERIENCE_SPEC.md` | 六组合首局、状态依赖选择、完整结果 | +| p91-ux-visual-designer | `docs/runtime/p9_1/UX_VISUAL_FLOW_SPEC.md` | 页面/状态/中文因果文案/横屏层级 | +| p91-runtime-architect | `docs/runtime/p9_1/RUNTIME_ARCHITECTURE_SPEC.md` | API、状态所有权、拆分次序、测试接缝 | +| p91-runtime-foundation-developer | shared runtime extraction | 行为等价、回归安全、并行写入面 | +| p91-domain-developer | `runtime/domain/p9_1/` | 任意组合的规则与事务 | +| p91-ui-developer | `runtime/ui/p9_1/` | 真实 Godot 页面与状态呈现 | +| p91-test-developer | P9.1 tests/capture helpers | 矩阵、触摸、布局、严格诊断 | +| p91-integration-developer | integration-owned shared files | 完整首局、全回归、H5 预览 | +| p91-independent-reviewer | read-only fixed revision | PASS/FAIL/INVALID | -## 本机环境 +## 本机资源事实 -| 能力 | 状态 | 说明 | -|---|---|---| -| Git/worktree | ✅ | clean main; two redesign worktrees will bind to the PMO baseline commit | -| codex 登录 | ✅ | codex-cli 0.146.0 | -| 开新会话可用路径 | ✅ | HAPI Hub local runner READY; spawn/inspect/message/catalog verified | -| Owner 可见性偏好 | 所有本批 worker/reviewer 均 HAPI 可见 | -| codex 线程工具 | 宿主未暴露 create_thread；使用正式 HAPI Hub spawn，不使用不可见 spawn_agent 冒充会话 | -| 监控机制登记 | R5 dual review active after exact ACKs; startup 90s then steady 600s | +- P9.0R10 product source: `41204acafa911a505f5b24a493aa032983e706d1`。 +- P9.0R10 closure: `f0b7d7b0234d8ad304f7b9d39fa99ded1dbfc27d`。 +- 当前没有本批次拥有的 Godot/Xcode/sign/upload 进程。 +- 其他项目进程不是本批资源，不能终止或归为本批失败。 -## 归档约定 +## 归档 -STATUS 由 queue/ledger 派生；完成任务按月归档，聊天不作为唯一事实源。 +先前 `recruit-line-redesign-02` 的完整状态保存在 +`.agent-taskgraph/archive/recruit-line-redesign-02/`，只作为历史证据，不参与 P9.1 状态计算。 diff --git a/.agent-taskgraph/ROLES.md b/.agent-taskgraph/ROLES.md index c414a4fb..d2bcd3ba 100644 --- a/.agent-taskgraph/ROLES.md +++ b/.agent-taskgraph/ROLES.md @@ -1,48 +1,22 @@ # 角色注册表 -> Team revision: `rev-13` +> Team revision: `rev-p9.1-1` | Role ID | 名称 | 生命周期 | 状态 | 核心职责 | 当前 Goal | Session ID | 最后更新 | |---|---|---|---|---|---|---|---| -| systems-designer | 系统与数值设计师 | persistent | assigned | 横向长队列、招募、战斗、经济、失败与保存合同 | task:ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR | 22caeb51-307e-41a6-af3e-875b8e37ba6b | 2026-08-11 | -| content-architect | 内容架构师 | persistent | reserved | 生成式成员、职业、特性、装备、敌人、Boss 与区域矩阵 | task:ABG-V2-CONTENT-OUTCOME-SYNC | HAPI 593d50d6 | 2026-08-11 | -| presentation-designer | 横屏体验与信息架构师 | persistent | assigned | 完整页面、状态、引导、输入、可访问性与 UI 语法 | task:ABG-V2-ONBOARDING-PRESENTATION-REPAIR | c5b75ce7-3baa-45aa-a037-c7f8491c68ee | 2026-08-11 | -| art-director-v2 | 角色场景与动效美术总监 | persistent | assigned | 模块化成员、职业、敌人、区域、动画与 VFX 生产规格 | task:ABG-V2-ART-SCHEMA-ORDER-REPAIR | 28cb1721-b66f-4fdf-9700-5ba5924761fa | 2026-08-11 | -| audio-director-v2 | 游戏音频总监 | persistent | assigned | 音乐、环境、SFX、语音、字幕与生命周期音频规格 | task:ABG-V2-AUDIO-COVERAGE-REPAIR | 00ed3a77-7d43-4e45-8dd0-a0e687775b5c | 2026-08-11 | -| design-validator-v2 | 设计验证工程师 | task-scoped | retired | 设计计数、ID、交叉引用、退役概念和覆盖率验证 | task:ABG-V2-CROSSDOC-VALIDATOR-H1 | 0eb4b8b3-4830-4ab1-bfbb-b5a643ce720d | 2026-08-11 | -| reference-researcher | 参考普查分析师 | persistent | assigned | 证据边界、结构经验和当前原创范围的一致性 | task:ABG-V2-REFERENCE-SCOPE-REPAIR | 8e9f4df2-06de-4660-953d-8aa3a9c31199 | 2026-08-11 | -| independent-reviewer-v2 | 独立设计审查员（R1 历史） | task-scoped | retired | 固定 revision 只读检查参考边界与新版一致性 | task:ABG-V2-REVIEW | HAPI 5161887f-fc11-4850-92b7-321ac8aa1bc2 | 2026-08-11 | -| systems-integrity-reviewer-v2 | 系统完整性审查员 | task-scoped | retired | 核心循环、战斗、经济、节奏、存档、失败与内容机械一致性 | task:ABG-V2-SYSTEMS-REVIEW-R2 | HAPI 7945e8ff-abe9-45ef-9f22-2838dacfa197 | 2026-08-11 | -| experience-production-reviewer-v2 | 体验与生产审查员 | task-scoped | retired | 页面可理解性、内容呈现、美术动画、音频、可访问性与生产可行性 | task:ABG-V2-EXPERIENCE-REVIEW-R2 | HAPI 8cfe0437-3a7e-4ac0-8b26-3477ef170654 | 2026-08-11 | -| systems-integrity-reviewer-r3 | 系统完整性独立复审员 R3 | task-scoped | retired | 固定版本核心循环、战斗、经济、节奏、存档、失败与机械内容一致性 | task:ABG-V2-SYSTEMS-REVIEW-R3 | afcafe40-d67b-4372-a63a-6cb94c2e6127 | 2026-08-11 | -| experience-production-reviewer-r3 | 体验与生产独立复审员 R3 | task-scoped | retired | 固定版本可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R3 | 44652407-b548-4f47-b78e-fca47657103f | 2026-08-11 | -| systems-integrity-reviewer-r4 | 系统完整性独立复审员 R4 | task-scoped | retired | 固定版本核心循环、战斗、经济、节奏、存档、失败与机械内容一致性 | task:ABG-V2-SYSTEMS-REVIEW-R4 | ba21ab5a-b78f-4c30-ad12-94b457dcec71 (archived) | 2026-08-11 | -| experience-production-reviewer-r4 | 体验与生产独立复审员 R4 | task-scoped | retired | 固定版本可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R4 | 0e8eec99-d95f-4b98-a9bf-cb5ac46433e1 (archived) | 2026-08-11 | -| run-creation-contract-repairer-r5 | 开局种子与事务合同修复员 | task-scoped | retired | 统一开局 seed、offer receipt、选择恢复与 current_run 原子提交 | task:ABG-V2-RUN-CREATION-TRANSACTION-REPAIR | 258f86c1-239c-4118-b782-55d7ed79c63b (archived) | 2026-08-11 | -| hidden-profession-art-repairer-r5 | 隐藏职业美术绑定修复员 | task-scoped | retired | 对齐 PF-H01..H06 父职业、轮廓、道具、动作与生产规格 | task:ABG-V2-HIDDEN-PROFESSION-ART-REPAIR | c32326b0-6952-46d2-b3b2-31347fa55ee4 (archived) | 2026-08-11 | -| design-validator-r5-h2 | R4 矛盾回归验证工程师 | task-scoped | retired | 为 seed/事务与隐藏职业美术绑定建立语义正负回归门 | task:ABG-V2-R4-VALIDATOR-H2 | 0ab9a014-89c7-4527-857c-6928349d842a (archived) | 2026-08-11 | -| design-validator-r5-h2-a2 | R4 矛盾回归验证工程师（替代尝试） | task-scoped | retired | 为 seed/事务与隐藏职业美术绑定建立语义正负回归门 | task:ABG-V2-R4-VALIDATOR-H2 | 2cf38e48-f8f9-4c61-adff-b822831c17c7 (archived) | 2026-08-11 | -| design-validator-r5-h2-a3 | R4 矛盾回归验证工程师（握手恢复） | task-scoped | retired | 为 seed/事务与隐藏职业美术绑定建立语义正负回归门 | task:ABG-V2-R4-VALIDATOR-H2 | 1546960c-baf0-4a36-bd71-b1c39bc577b7 (archived) | 2026-08-11 | -| systems-integrity-reviewer-r5 | 系统完整性独立复审员 R5 | task-scoped | retired | 固定源核心循环、战斗、经济、节奏、存档、失败与机械内容一致性 | task:ABG-V2-SYSTEMS-REVIEW-R5 | a0ea4f06-a73d-47c4-bc0d-dad36c6d735d (archived) | 2026-08-11 | -| experience-production-reviewer-r5 | 体验与生产独立复审员 R5 | task-scoped | retired | 固定源可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R5 | c52000b5-2021-48cb-8d68-299f36876b8c (archived INVALID) | 2026-08-11 | -| experience-production-reviewer-r5-a2 | 体验与生产独立复审员 R5 替代 | task-scoped | retired | 固定源可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R5 | 2392ec4d-83e8-4df4-b06e-0a7d3fa8167b (archived) | 2026-08-11 | -| region-art-binding-repairer-r6 | 区域内容到美术生产绑定修复员 | task-scoped | assigned | 对齐 RG-01..08 与 ART-REG-01..08，并补语义回归门 | task:ABG-V2-REGION-ART-BINDING-R6 | c63c1b13-8f65-479c-ad65-b2c998713a2d | 2026-08-11 | -| final-prototype-builder-r6 | 最终横屏 HTML 原型制作员 | task-scoped | assigned | 将当前 50 页面、8 区域和完整目录实现为 Owner 可审的 Gallery 原型 | task:ABG-V2-FINAL-PROTOTYPE-R6 | 397e65db-795c-46c4-af7a-5eccf23ab56e | 2026-08-11 | +| p91-game-designer | P9.1 游戏策划 | persistent | reserved | 六组合首局玩法、取舍、节奏和结果 | task:ABG-P9.1-GAME-DESIGN | PENDING | 2026-08-13 | +| p91-ux-visual-designer | P9.1 UX/视觉设计师 | persistent | reserved | 横屏页面、状态、中文因果表达与动效 | task:ABG-P9.1-UX-VISUAL | PENDING | 2026-08-13 | +| p91-runtime-architect | P9.1 技术架构师 | persistent | reserved | 运行时拆分、API、状态所有权和迁移 | task:ABG-P9.1-TECH-ARCH | PENDING | 2026-08-13 | +| p91-runtime-foundation-developer | 运行时基础开发 | persistent | available | 行为等价拆分与并行扩展边界 | PENDING | PENDING | 2026-08-13 | +| p91-domain-developer | 领域逻辑开发 | persistent | available | 六组合领域状态和事务 | PENDING | PENDING | 2026-08-13 | +| p91-ui-developer | Godot UI 开发 | persistent | available | P9.1 真实页面与反馈 | PENDING | PENDING | 2026-08-13 | +| p91-test-developer | 测试与证据开发 | persistent | available | 矩阵、触摸、布局与捕获工具 | PENDING | PENDING | 2026-08-13 | +| p91-integration-developer | 运行时集成开发 | persistent | available | 接入共享入口与完整回归 | PENDING | PENDING | 2026-08-13 | +| p91-independent-reviewer | P9.1 独立 reviewer | task-scoped | available | 固定 revision 只读审查 | PENDING | PENDING | 2026-08-13 | +| p91-h5-publisher | P9.1 H5 发布员 | task-scoped | available | 固定源 Web 导出与 Gallery 发布 | PENDING | PENDING | 2026-08-13 | ## 编制变更历史 | 时间 | Team revision | 动作 | Owner 授权 | 结果 | |---|---|---|---|---| -| 2026-08-10 | rev-1 | 旧固定角色/三路线设计批次 | Owner 初始授权 | Owner 后续停止，历史保留 | -| 2026-08-10 | rev-2 | 两个写入会话并行 + 一个 Owner 手动开启的可见 reviewer | Owner 明确允许多个会话继续 | 等待 worker runtime 验证 | -| 2026-08-10 | rev-3 | 新增 UX、美术、音频、设计验证四个互斥写入会话 | Owner：`可以再多几个会话，10个以内我觉得都OK` | 执行并发扩大到六个 worker；reviewer 后置待命 | -| 2026-08-11 | rev-4 | R1 失败修复后拆为系统完整性与体验生产两个并行只读 reviewer | Owner：`可以再多几个会话，10个以内我觉得都OK` | 六个写入会话已交候选；新增两个 reviewer，总量不超过 10 | -| 2026-08-11 | rev-5 | Experience R3 FAIL 后复用 UX/Art/Audio/Validator 会话并新增一个 Reference 修复会话 | 同一 Owner 并发授权；不扩大产品范围 | 五个互斥修复 Goal；Systems R3 继续只读固定版本 | -| 2026-08-11 | rev-6 | H1 集成全绿后新增两个全新 R4 独立 reviewer | Owner 已授权最多 10 个可见会话与 fresh 双审 | 两个 reviewer 只读同一固定源 `db932cf`；R1-R3 结论只作历史 | -| 2026-08-11 | rev-7 | R4 双 FAIL 后新增两个互斥 task-scoped 修复角色 | 既有 Owner 并发授权；失败路由不扩大产品范围 | seed/transaction 三文档与 hidden-art 单文档并行，之后 validator H2 + fresh R5 | -| 2026-08-11 | rev-8 | R4 两项修复整合后新增 task-scoped H2 validator | graph-r10 已批准失败路由；既有 HAPI profile 授权 | 只改验证合同与脚本；完成后 fresh R5 双审 | -| 2026-08-11 | rev-9 | H2 attempt-1 身份流程无效后替换 task-scoped validator | Owner：`OK的啊继续吧`，明确批准 `1/1 -> 2/2` replacement attempt | 旧角色保持 retired；新角色独占 attempt-2，两文件范围不变 | -| 2026-08-11 | rev-10 | 将零编辑 pre-dispatch 身份失败改为 PMO 自动恢复 | Owner：`你继续啊，为什么一直要我授权` | 新建握手恢复角色；两阶段 dispatch；不消耗产品 Goal 尝试 | -| 2026-08-11 | rev-11 | H2 整合后新增两个 fresh R5 独立 reviewer | 既有 Owner 最多 10 个可见会话及双 reviewer 授权 | 两个只读角色并行审同一固定源；双 PASS 才进入 synthesis | -| 2026-08-11 | rev-12 | Experience R5 首次 harness INVALID 后自动替换 | PROJECT pre-dispatch/reviewer environment recovery policy | 原 reviewer 归档；fresh a2 仍只读同一固定源 | -| 2026-08-11 | rev-13 | Experience R5 Gate 8 后并行区域绑定修复与最终 HTML 制作 | Owner 要求尽快完成完整策划、美术与可查看输出；既有最多 10 个可见会话授权 | 两个写入域互斥；不进入引擎开发或最终媒体生产 | +| 2026-08-13 | rev-p9.1-1 | 建立三专家波次与后续分岗开发 | Owner 明确要求 agent graph 并行安排团队 | 首批三角色 reserved，等待 HAPI runtime 验证和 ACK | diff --git a/.agent-taskgraph/STATUS.md b/.agent-taskgraph/STATUS.md index a684bfd2..366d95a8 100644 --- a/.agent-taskgraph/STATUS.md +++ b/.agent-taskgraph/STATUS.md @@ -2,44 +2,10 @@ | 任务 ID | 标题 | 状态 | 负责人 | 轮次 | 最后更新 | 卡点 | |---|---|---|---|---|---|---| -| ABG-V2-CORE-SYSTEMS | 横向长队列核心系统重构 | review | systems-designer | 2/2 | 2026-08-11T00:47:43+08:00 | repair `e598eed` 已整合为 `9be1927`；等待双 reviewer | -| ABG-V2-CONTENT-MATRIX | 生成式招募与完整内容矩阵 | review | content-architect | 1/2 | 2026-08-10T23:41:16+08:00 | candidates `4e07a91` + `6a41189` 已整合；ACK 到达顺序偏晚已记录 | -| ABG-V2-UX-SCREENS | 横屏 UX 与完整页面状态 | review | presentation-designer | 2/2 | 2026-08-11T00:47:43+08:00 | repair `c64b99e` 已整合为 `ec4863a`；页面 50、绑定 33 | -| ABG-V2-ART-DIRECTION | 模块化角色、场景、动画与 VFX 规格 | review | art-director-v2 | 1/2 | 2026-08-10T23:40:01+08:00 | candidate `ca4d3df` 已整合 | -| ABG-V2-AUDIO-DIRECTION | 音乐、环境、SFX 与语音规格 | review | audio-director-v2 | 2/2 | 2026-08-11T00:47:43+08:00 | repair `ff4101a` 已整合为 `f0095af`；17 个绑定已关闭 | -| ABG-V2-DESIGN-VALIDATION | 完整设计机械验证合同与工具 | review | design-validator-v2 | 2/2 | 2026-08-11T00:47:43+08:00 | repair `e7f80d0` 已整合为 `83bba23`；self-test/full 均通过 | -| ABG-V2-REVIEW | 独立只读完整设计审查 | failed | independent-reviewer-v2 | 1/1 | 2026-08-11T00:34:00+08:00 | FAIL：最后成员解雇可形成未定义 0 人 Market；修复后需新 reviewer | -| ABG-V2-SYSTEMS-REVIEW-R2 | 系统、经济、战斗与存档完整性审查 | failed | systems-integrity-reviewer-v2 | 1/1 | 2026-08-11T01:10:00+08:00 | FAIL Gate 5：PF-A04 依赖未定义 hit/miss/evasion 系统 | -| ABG-V2-EXPERIENCE-REVIEW-R2 | UX、内容呈现与美术音频生产审查 | failed | experience-production-reviewer-v2 | 1/1 | 2026-08-11T01:10:00+08:00 | INVALID：PMO context 引用了固定 revision 中不存在的文件名 | -| ABG-V2-BATTLE-OUTCOME-REPAIR | 确定性战斗结果词汇修复 | review | systems-designer | 1/1 | 2026-08-11T08:25:23+08:00 | candidate `5a05ffc` 整合为 `f2a3f3c`；三项验证通过 | -| ABG-V2-CONTENT-OUTCOME-SYNC | 内容结果词汇同步 | review | content-architect | 1/1 | 2026-08-11T09:14:42+08:00 | `d5fd2fc` -> `edbd281` 已集成冻结；等待 repaired 双审 | -| ABG-V2-ART-OUTCOME-SYNC | 美术与 VFX 结果词汇同步 | review | art-director-v2 | 1/1 | 2026-08-11T09:14:42+08:00 | `8c8c1b7` -> `f8d5bfe` 已集成冻结；新任务只修 schema/order | -| ABG-V2-AUDIO-OUTCOME-SYNC | 音频结果词汇同步 | review | audio-director-v2 | 1/1 | 2026-08-11T09:14:42+08:00 | `ea69bc7` -> `404e996` 已集成冻结；新任务只修 coverage | -| ABG-V2-SYSTEMS-REVIEW-R3 | 系统完整性固定版本复审 | failed | systems-integrity-reviewer-r3 | 1/1 | 2026-08-11T09:36:00+08:00 | FAIL Gate 6：缺五个可执行固定种子反支配 fixture | -| ABG-V2-EXPERIENCE-REVIEW-R3 | 体验与生产固定版本复审 | failed | experience-production-reviewer-r3 | 1/1 | 2026-08-11T09:14:42+08:00 | FAIL：缺 provisional choose-two 页面；另有方向/schema/audio/reference 矛盾 | -| ABG-V2-ONBOARDING-PRESENTATION-REPAIR | 开局选择、引导与队列方向修复 | review | presentation-designer | 1/1 | 2026-08-11T10:04:32+08:00 | `388d90b` -> `14ea46e` 已整合；待 R4 | -| ABG-V2-ART-SCHEMA-ORDER-REPAIR | Recruit 美术 schema 与队列方向修复 | review | art-director-v2 | 1/1 | 2026-08-11T10:04:32+08:00 | `7f8d4b4` -> `03547a7` 已整合；late ACK 偏差保留；待 R4 | -| ABG-V2-AUDIO-COVERAGE-REPAIR | OUT-006 与 50 页面音频覆盖修复 | review | audio-director-v2 | 1/1 | 2026-08-11T10:04:32+08:00 | `9f63e46` -> `7071887` 已整合；late ACK 偏差保留；待 R4 | -| ABG-V2-REFERENCE-SCOPE-REPAIR | 参考经验中的旧固定角色范围修复 | review | reference-researcher | 1/1 | 2026-08-11T10:04:32+08:00 | `565ba2e` -> `96ea49a` 已整合；待 R4 | -| ABG-V2-CROSSDOC-VALIDATOR-REPAIR | R3 跨文档矛盾回归门 | failed | design-validator-v2 | 1/1 | 2026-08-11T10:17:06+08:00 | INVALID：真实整合文档出现 3 个 parser 假阴性；转 H1 | -| ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR | 固定种子构筑反支配模拟修复 | review | systems-designer | 1/1 | 2026-08-11T10:17:06+08:00 | `cb53794` -> `b65646c` 已整合；模拟与 hashes 通过；待 H1/R4 | -| ABG-V2-CROSSDOC-VALIDATOR-H1 | 真实文档 validator 集成修复 | review | design-validator-v2 | 1/1 | 2026-08-11T10:44:55+08:00 | `49e128f` -> `bcb2931` 已整合；20-case/self/full/JSON 与反支配回放全绿；待 R4 双审 | -| ABG-V2-RUN-CREATION-TRANSACTION-REPAIR | 开局 seed 与可恢复事务修复 | review | run-creation-contract-repairer-r5 | 2/2 | 2026-08-11T11:58:55+08:00 | `ad0a57d` -> `bba0c43` 已整合；等待 H2 + fresh Systems R5 | -| ABG-V2-HIDDEN-PROFESSION-ART-REPAIR | 隐藏职业内容到美术绑定修复 | review | hidden-profession-art-repairer-r5 | 2/2 | 2026-08-11T11:58:55+08:00 | `fef9057` -> `8102bd2` 已整合；`c558d19` 禁止；等待 H2 + fresh Experience R5 | -| ABG-V2-R4-VALIDATOR-H2 | R4 seed/事务与隐藏职业绑定回归门 | done | design-validator-r5-h2-a3 | product 1/2; dispatch recovery 3 | 2026-08-11T13:06:00+08:00 | candidate `fb20b66` 整合为 `7dc584d`；24 cases/errors=0/JSON/5 seeds 通过；等待 R5 双审 | -| ABG-V2-SYSTEMS-REVIEW-R5 | 系统完整性固定源复审 | failed | systems-integrity-reviewer-r5 | 1/1 | 2026-08-11T13:25:00+08:00 | reviewer Gate 1-8 PASS；Gate 9 使用错误时长 Goal，PMO 分类 INVALID_REVIEW_CONTRACT；Gate 10 未审 | -| ABG-V2-EXPERIENCE-REVIEW-R5 | 体验与生产固定源复审 | failed | experience-production-reviewer-r5-a2 | product judgment 1/1; environment recovery 2 | 2026-08-11T13:48:00+08:00 | Gates 1-7 PASS；Gate 8 FAIL：ART-REG 未绑定 RG 且序号简报冲突；9-10 未提升 | -| ABG-V2-REGION-ART-BINDING-R6 | 区域美术绑定收口 | active | region-art-binding-repairer-r6 | 1/1 | 2026-08-11T13:57:00+08:00 | exact rev-13 ACK verified；发送完整 Goal | -| ABG-V2-FINAL-PROTOTYPE-R6 | 最终横屏 HTML/Gallery 原型 | active | final-prototype-builder-r6 | 1/1 | 2026-08-11T13:57:00+08:00 | exact rev-13 ACK verified；发送完整 Goal | - -## 批次 - -- `recruit-line-redesign-02` -- Frozen spec: `design-spec-r3` -- Runtime: HAPI visible Codex `gpt-5.6-sol / xhigh / yolo` -- 并发上限：6；H2 已整合；正在冻结同一 R5 source 并准备两个全新只读 reviewer。 -- HTML、最终美术/音频生成、运行时、打包和发布均未授权。 ## 当前批次总结 -- R4 两项产品修复及 H2 回归门已整合。下一步只对同一不可变 R5 source 做 Systems/Experience 双审；只有双 PASS 才进入 PMO synthesis 与 Owner 设计门。HTML 与正式开发继续冻结。 +- 批次：`abg-p9-1-continuous-development-01`。 +- 当前 Gate：三专家任务进入 inbox；HAPI 控制面 VERIFIED + exact ACK 后才能 active。 +- 开发 Gate：三份专家规格由 PMO 汇合并冻结前，所有 runtime 实现节点保持 pending。 +- 结束 Gate：独立 reviewer PASS 后发布新 H5，Owner 再判断理解、乐趣和继续意愿。 diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/DECISIONS.md b/.agent-taskgraph/archive/recruit-line-redesign-02/DECISIONS.md new file mode 100644 index 00000000..c0bd146c --- /dev/null +++ b/.agent-taskgraph/archive/recruit-line-redesign-02/DECISIONS.md @@ -0,0 +1,164 @@ +# Decisions + +## 2026-08-10 Product Direction + +- Original product: `Aetherbound Guild / 裂脉远征团`. +- Familiar, readable magical guild and job fantasy; no Dragon Quest or reference-game expression is copied. +- Complete scope: 30 named characters, 12 base jobs, 24 advanced jobs, 100 enemy identities, 8 chapters and 320 equipment/items. +- Target playtime: 25-35 hour first campaign, 60-80 hour build completion and roughly 100 hour full collection/high difficulty. +- Landscape 16:9; full 2D Japanese hand-drawn direction using layered rigs, equipment layers, authored key poses and unique portraits/illustrations. +- Formal implementation begins only after the complete design package, HTML gallery and interactive design prototype pass Owner review. + +## 2026-08-10 Execution Decision + +- Use Agent TaskGraph protocol 0.8.0-beta.10. +- Run three parallel non-overlapping design workers, then presentation, integration and an independent reviewer. +- All worker and reviewer sessions use Codex `gpt-5.6-sol`, `xhigh`, project-level `yolo`. +- Native `wait_agent` is the monitoring mechanism; no shell sleep or passive chat waiting. + +## 2026-08-10 R2 Runtime Recovery + +- R1 correctly stopped on a PMO worktree-path mismatch. Goal paths were corrected, context revisions advanced to 2, and dispatch IDs advanced to r2. +- Because the original collaboration follow-up did not yield a valid recoverable ACK, the same local Codex runtime was relaunched through `codex exec` with unchanged `gpt-5.6-sol`, `xhigh`, `yolo` and isolated worktrees. +- CLI session/JSONL IDs are the r2 identity evidence. R1 draft output is not accepted unless the r2 worker inspects, completes, commits and submits it. +# ABG-SYSTEM-REPAIR + +| 时间 | 决策 | 决策者 | 影响 | +|---|---|---|---| +| 2026-08-10 | 接受 HAPI reviewer 对 `22b21f8` 的 FAIL，批准同一 systems-designer 角色进行一次有界前向修复；为保持活跃 batch 的已确认 runtime 与 validator 一致，worker 复用本批 Codex native profile，修复后的独立 reviewer 仍为 HAPI 可见；固定 `gpt-5.6-sol / xhigh / yolo` | Owner：`继续` | graph-r2 新增 `system_design_repair`；Frozen 产品范围不变；表现设计依赖改为修复节点 | + +# Stop External Orchestration + +| 时间 | 决策 | 决策者 | 影响 | +|---|---|---|---| +| 2026-08-10 | 停止本批 AgentGraph、所有外部 worker/reviewer 和后台监控，后续在当前 Owner 会话直接整合与修复 | Owner：`停止了吧，就在当前会话进行吧` | Batch 01 关闭；候选提交保留；无会话继续执行；不再使用 queue 状态驱动本轮设计工作 | + +# Recruit-Line Redesign Batch 02 + +| 时间 | 决策 | 决策者 | 影响 | +|---|---|---|---| +| 2026-08-10 | 以参考游戏已验证的横向队列、持续招募、装备构筑和自动战斗为玩法骨架；删除固定 30 名剧情角色、关系系统、固定六人、三路线和战中 Directive | Owner | 旧产品合同作废，必须前向重写 | +| 2026-08-10 | 完整原创规模采用 12 基础 + 24 进阶 + 6 隐藏职业、36 特性、320 装备、60 神器、100 普通/精英敌人、16 Boss、8 区域 | Owner：`对可以这样设计` | design-spec-r2 冻结 | +| 2026-08-10 | 允许开启多个可见会话；两个无冲突写入 worker 并行，Owner 手动开启的 `审计 Aetherbound 复刻差距` 会话保留为后置只读 reviewer | Owner | HAPI visible Codex gpt-5.6-sol/xhigh/yolo；UI 工作等待系统与内容验收 | + +# Recruit-Line Redesign Batch 02 Parallel Expansion + +| 时间 | 决策 | 决策者 | 影响 | +|---|---|---|---| +| 2026-08-10 | 本批可在真实职责和无写入冲突前提下扩展到 10 个以内可见会话；先增设 UX、美术、音频与设计验证四条设计线 | Owner：`可以再多几个会话，10个以内我觉得都OK` | design-spec-r3 / graph-r4；六个 worker 并行，一个独立 reviewer 后置；HTML 与运行时仍冻结 | + +# Recruit-Line Redesign Batch 02 Dual Review + +| 时间 | 决策 | 决策者 | 影响 | +|---|---|---|---| +| 2026-08-11 | 六条设计线完成后，不再新增重复写作者；新增系统完整性与体验生产两个并行只读 reviewer，PMO 汇总后进入 Owner Gate | Owner：`可以再多几个会话，10个以内我觉得都OK` | Team rev-4 / graph-r5；项目相关可见会话总量不超过 10；HTML、引擎、最终媒体、打包与发布继续冻结 | + +# Recruit-Line Redesign R2 Review Routing + +| 时间 | 决策 | 决策者 | 影响 | +|---|---|---|---| +| 2026-08-11 | 接受 Systems R2 对 `7054fe6` 的 Gate 5 FAIL；保留既有“无隐藏 miss chance”，先定义确定性 action outcome，再并行同步 content/art/audio，不新增 accuracy/evasion 概率系统 | PMO，依据 Frozen 可理解性与既有战斗权威 | graph-r6；复用原 worker 角色；新固定 revision 后必须使用两个 fresh reviewer | +| 2026-08-11 | Experience R2 因 context 引用了不存在的 `REFERENCE_ANALYSIS_AND_LESSONS.md` 判为 INVALID；不作产品判断、不复用会话 | PMO reviewer protocol | 修正未来 R3 context 为现存 `REFERENCE_RUNTIME_CENSUS.md` 与 `TRANSFERABLE_STRUCTURE_LESSONS.md` | +# 2026-08-11 - Experience R3 Repair Routing + +- Decision: accept `ABG-V2-EXPERIENCE-REVIEW-R3 FAIL` as a product-design failure on immutable `d454120`. +- Reason: run creation has no provisional choose-two presentation state, party-line direction conflicts, recruit content/art schemas lack a crosswalk, audio omits `OUT-006` and miscounts screens, and the lessons authority retains fixed-cast scope. +- Route: reuse the persistent UX, Art, Audio and Validator sessions with disjoint writes; add one visible Reference repair session. Systems R3 continues independently on the immutable source. +- Frozen: no HTML, runtime, final media, packaging or release. Repaired documents require a new immutable revision and fresh experience R4 review. + +# 2026-08-11 - Systems R3 Gate 6 Repair Routing + +- Decision: accept the immutable `d454120` Systems R3 FAIL and route only the missing fixed-seed anti-dominance proof to the existing Systems role. +- Reason: battle, transaction, formula, pacing and solvency checks passed; the earliest failure is absent executable seed fixtures and policy outcomes. +- Route: one new fixture JSON, one simulation tool and the economy authority only; later integration creates fresh Systems R4 and Experience R4 reviews. +- Frozen: no stochastic accuracy/evasion, unrelated rebalance, content expansion, HTML, runtime, media, packaging or release. + +# 2026-08-11 - R3 Repair Runtime Reconciliation + +- Art candidate `7f8d4b4` and Audio candidate `9f63e46` were edited, committed and pushed before PMO observed their exact standalone rev-5 identity ACKs. Their later ACKs reconcile identity but do not erase the process-order deviation or independently accept the product changes. +- Reference candidate `565ba2e` reached its legal terminal with clean/upstream evidence. Systems HAPI `22caeb51-307e-41a6-af3e-875b8e37ba6b` returned the exact ACK before editing and is active. +- Current TaskGraph state validation now accepts the six current tasks' dispatch/runtime/ACK fields. Remaining errors are pre-existing review-role concurrency/history debt; the installed graph validator also rejects the repository's already-approved `fork` node kind. Do not expand this bounded repair into a wholesale legacy-ledger or graph-schema migration. +- UX `388d90b`, Art `7f8d4b4`, Audio `9f63e46`, Reference `565ba2e`, and Validator `ec5e234` passed PMO write-scope/evidence inspection and were integrated without conflict as `14ea46e`, `03547a7`, `7071887`, `96ea49a`, and `8a6f0c8`. This is integration, not R4 or Owner acceptance. +- Systems `cb53794` was integrated as `b65646c`; anti-dominance self-test and canonical replay pass. Combined validation then exposed three validator-only false negatives against real repaired prose/tables while its generated fixture remained green. R4 is blocked on a two-file H1 parser/regression repair; product authorities remain frozen. + +# 2026-08-11 - H1 Integration Acceptance And R4 Freeze + +- H1 candidate `49e128f` changes only the validation contract and executable; PMO integrated it as `bcb2931` and independently reran the combined validator and anti-dominance gates. +- Text and JSON validation return zero diagnostics, the validator self-test remains 20 cases, the five contradiction mutations remain rejected, and the five fixed policy seeds preserve a non-dominant winner spread. Result SHA-256: `17154371d4b93035296d1219f012f381af2e10a5f2c1b3f66bcfcd7ad4d85352`. +- Create one PMO documentation closure and use that exact pushed revision as the sole immutable source for fresh Systems R4 and Experience R4 reviewers. No R1-R3 partial judgment may be reused as PASS. +- HTML, engine work, final media, packaging and release remain frozen. Dual reviewer PASS leads only to PMO synthesis and the Owner design gate. + +# 2026-08-11 - R4 Review Failure Routing + +- Accept Systems R4 FAIL on immutable `db932cf`: provisional generated Recruits are selected before the only declared run seed exists, so offer reproduction, selection receipts and crash retry are undefined. +- Accept Experience R4 FAIL on the same immutable source: `PF-H01..H06` content mechanics/production briefs and art parent/silhouette/motion bindings are inconsistent, with unresolved-binding language still present. +- Route two disjoint minimal repairs in parallel: seed/transaction authority across the exact product/content documents, and hidden-profession binding inside the art catalog only. Do not broaden balance, content count, UI, audio, HTML, runtime or media scope. +- After integration, add deterministic validator negatives for both R4 contradiction classes and create a new fixed revision. Only fresh dual reviewer PASS may reach the Owner gate. + +# 2026-08-11 - R5 Repair Attempt-1 Identity Invalidation + +- Both attempt-1 repair sessions edited before returning the exact standalone revision-bound `IDENTITY_READY`; PMO STOP messages did not prevent later work. Both sessions are archived and their runtime evidence is retained as `runtime-evidence-attempt1-invalid.json`. +- The run-creation attempt-1 dirty worktree is preserved unchanged as invalid history. Hidden-art commit `c558d19c1c96a07ab9e7caa23f495875126bc113` is explicitly forbidden from integration. +- One final attempt is authorized under the existing failure route: fresh clean `-a2` worktrees at `0b0f6ad`, context revision 2, dispatch attempt 2, exact identity ACK before any edit. A second identity-process failure stops the branch and requires Owner action. + +# 2026-08-11 - R4 Repair Integration And H2 Route + +- Attempt-2 workers returned exact identity ACKs before any edit, stayed within the three-file transaction scope and one-file hidden-art scope, pushed clean candidates `ad0a57d` and `fef9057`, and were archived after their legal terminals. +- PMO integrated them as `bba0c43` and `8102bd2`. This closes the two bounded repair nodes only; it does not grant complete design or Owner acceptance. +- Continue to the approved `v2_r4_validator_h2` node, then freeze one new source for fresh Systems R5 and Experience R5 reviews. HTML, engine, media production, packaging and release remain frozen. + +# 2026-08-11 - Validator H2 Staffing + +- Add one fresh task-scoped `design-validator-r5-h2` role under Team rev-8 rather than reviving the retired H1 validator identity. +- H2 owns only the validation contract and executable. It must preserve the 20 existing self-test cases and add exactly four isolated R4 contradiction cases before any R5 reviewer is created. + +# 2026-08-11 - Validator H2 Attempt-1 Invalid + +- HAPI `0ab9a014` passed runtime verification but did not emit the required exact standalone identity ACK as its first agent response. PMO observed a still-clean worktree, sent STOP, archived the session and classified the attempt `INVALID_IDENTITY_PROCESS`. +- No H2 edit, test, commit, push or judgment exists. R4 product repairs remain integrated at `8102bd2`; R5 dual review remains blocked on H2. +- The graph/Goal limit is 1/1. A fresh attempt requires Owner authorization to advance the dispatch/context and max-attempt boundary. + +# 2026-08-11 - Validator H2 Attempt-2 Recovery Authorized + +- Owner explicitly answered `OK的啊继续吧`, authorizing the H2 attempt boundary to advance from `1/1` to `2/2` and a fresh replacement session. +- Retain attempt-1 files and runtime evidence as immutable INVALID identity-process history. The retired `design-validator-r5-h2` role and archived HAPI `0ab9a014` must not be reused. +- Add task-scoped replacement role `design-validator-r5-h2-a2` under Team `rev-9`, a new worktree/branch, context revision 2 and dispatch attempt 2. Exact standalone `IDENTITY_READY` remains a hard pre-edit gate. +- Product scope and acceptance do not change: only the validation contract and executable may change, exactly 24 self-tests are required, and fresh R5 dual review remains downstream. HTML, engine, media, packaging and release stay frozen. + +# 2026-08-11 - Validator H2 Attempt-2 Identity Invalid And Exhausted + +- HAPI `2cf38e48-f8f9-4c61-adff-b822831c17c7` passed pre-dispatch runtime verification and received the generated rev-9 bootstrap. Its first agent response was an analysis sentence rather than the exact standalone `IDENTITY_READY`. +- The real 90-second HAPI timer/observe cycle (`functions.wait` cell 1986) exposed the mismatch. PMO sent STOP, confirmed the fresh worktree remained clean at `b0e3475`, and archived the session before any validator edit, test, commit or push. +- Classify attempt-2 as `INVALID_IDENTITY_PROCESS`; it makes no product or validator judgment. The H2 `2/2` limit is exhausted, so R5 reviewers remain uncreated and the queue stops for a new Owner decision. + +# 2026-08-11 - Pre-dispatch Recovery Does Not Consume Product Attempts + +- Owner directed PMO to continue without repeatedly requesting authorization for identical process recovery. +- Correct the earlier accounting: a verified HAPI session that fails identity before ACK, stays outside `active`, and leaves the Goal worktree clean with zero edits/tests/commits does not consume `graph.max_attempts`. Both prior H2 sessions are retained as INVALID process history; H2 product execution remains `0/2`. +- PMO may automatically create another fresh session when runtime, machine, model, effort, permission, visibility, write scope, cost and Human Gates are unchanged. Any product edit, scope/cost/permission expansion, paid action, packaging or release boundary still requires the normal gate. +- Recovery 3 uses a two-stage dispatch: first message carries only Role/Team/Goal/Context/continuity identity and requests the exact ACK; after ACK is observed and recorded, PMO sends the rendered full Goal bootstrap. + +# 2026-08-11 - Validator H2 Integrated And R5 Review Unblocked + +- Recovery-3 HAPI `1546960c-baf0-4a36-bd71-b1c39bc577b7` returned the exact rev-10 ACK before any read/edit, then completed the bounded two-file Goal. +- Candidate `fb20b663df97bb685887e5bef9bb6a3256fa7580` changes only the validation contract and executable. It reports `cases=24`, integrated `errors=0`, byte-identical JSON SHA-256 `4feb0666acc639a163fefadda40a4cda3a6e5caf0db58fa01432c773ec8d6e38`, five anti-dominance seeds and all four isolated R4 diagnostics. +- PMO confirmed exact scope, preserved prior gates, integrated the candidate as `7dc584d62fc4323746c927f5dc370e77f992f0c7`, pushed clean/upstream and archived the worker. +- Freeze one new documentation closure as the shared immutable R5 source. Create fresh Systems R5 and Experience R5 read-only reviewers; neither may inherit R4 PASS/FAIL as current judgment. + +# 2026-08-11 - R5 Dual Independent Review Staffing + +- Use closure `69f6715b6f47b507fb07436ee1edf77df7e5adab` as the one immutable review source containing integrated H2 `7dc584d` and its PMO closure. +- Add fresh task-scoped Systems R5 and Experience R5 reviewer roles under Team rev-11. Both use visible HAPI Codex `gpt-5.6-sol / xhigh / yolo`, separate worktrees and zero repository writes. +- Systems reruns all gates from zero with the repaired seed/draft/promotion transaction. Experience reruns all gates from zero with exact hidden-profession production bindings. R4 findings are history, not current PASS. +- A PASS from one reviewer cannot compensate for FAIL/INVALID from the other. Only dual PASS proceeds to PMO synthesis and Owner design review. + +# 2026-08-11 - Experience R5 Environment Recovery + +- Experience R5 attempt-1 used zsh special variable `path` in its first archive-verification harness and failed before source judgment, then continued after PMO STOP. Classify the whole attempt INVALID and archive HAPI `c52000b5-2021-48cb-8d68-299f36876b8c`. +- Its assigned worktree remained clean at `e4ef2b0`; no product judgment or repository write is reusable. Automatically create one fresh replacement reviewer on the same immutable source and unchanged read-only Goal. + +# 2026-08-11 - Final Design Package Closure + +- Owner explicitly prioritizes fast completion of the complete pre-development product package and final reviewable output, not engine implementation. +- Preserve Experience R5 Gates 1-7 PASS and repair only its earliest real blocker: exact `RG-01..08` to `ART-REG-01..08` production mapping plus validator regression coverage. +- In parallel, replace the superseded 82-screen coverage audit with the current-authority 50-state landscape HTML prototype. No engine, bulk final media, packaging or release work is authorized. diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/PROJECT.md b/.agent-taskgraph/archive/recruit-line-redesign-02/PROJECT.md new file mode 100644 index 00000000..2f2867e5 --- /dev/null +++ b/.agent-taskgraph/archive/recruit-line-redesign-02/PROJECT.md @@ -0,0 +1,85 @@ +# 项目档案：Aetherbound Guild + +## 是什么 + +`Aetherbound Guild / 裂脉远征团` 是原创横屏 2D 日系手绘魔法公会远征策略游戏。当前只允许完整开发前设计和 HTML 原型；正式引擎开发冻结。 + +## 技术栈与目录结构 + +当前无运行时引擎。Markdown/JSON/CSV 是设计权威，HTML/CSS/JS 只用于设计原型。`docs/reference`、`docs/product`、`docs/content`、`docs/presentation`、`docs/prototype`、`docs/reviews` 按职责隔离。 + +## 硬性规范 + +- 授权商业参考只提炼结构，不复制名称、文案、公式、布局、代码、图片或音频。 +- 完整设计包独立审查和 Owner 审核前，禁止运行时代码、引擎项目、批量最终资产、打包和发布。 +- PMO 独占 `.agent-taskgraph/` 与根共享文件；并行 worker 只写 Goal 指定目录。 +- 每条内容必须有机械作用、交互、进度位置、表现、资产/音频和验证字段；禁止 filler。 + +## 需求与执行授权策略 + +| 配置 | 项目选择 | +|---|---| +| Agent TaskGraph 协议版本 | 0.8.0-beta.10（本活跃批次冻结；安装版 beta.12 只提供兼容 helper，不迁移协议） | +| Source baseline | READY: `/Users/wusumac/Documents/game/aetherbound_guild` at `ac0dec25848dbc6fe54f8de07c8396b01d8d3a89`, branch `main`, clean; redesign worktrees created only after the PMO baseline commit | +| 单任务快速路径 | 允许；本批不适用 | +| 复杂任务规格冻结 | `.agent-taskgraph/spec.md` design-spec-r3 已由 Owner 冻结 | +| 自主创建可见会话 | recruit-line-redesign-02 最多 10 个项目相关可见会话；完成六条设计线后新增两个独立 reviewer 已授权 | +| 派发预览授权 | 本批固定 HAPI 可见 Codex gpt-5.6-sol / xhigh / yolo 已授权 | +| 编制变更授权 | Owner 允许本批总会话数不超过 10；双 reviewer 分工已批准，仍须只读隔离、无写入冲突与真实 Goal | +| 默认 Human Gates | 依赖安装、删除、付费扩量、正式开发、打包、发布 | +| 已启用可选适配器 | HAPI：本批两个 worker 与 reviewer 均须用户可见、可接管 | +| Runtime 证据策略 | spawn 后验证线程、worktree、model、effort、permission 和 ACK；不匹配保持 inbox | +| 上下文模式 | balanced | +| Context 必读项上限 | 8 | + +## Execution profile（派发硬门） + +| Execution profile | Confirmed value | +|---|---| +| Execution profile status | CONFIRMED | +| Execution profile confirmed by/at | Owner / 2026-08-10T22:38:28+08:00 / recruit-line-redesign-02 | +| Runtime choice confirmed by/at | Owner / 2026-08-10 / multiple visible HAPI sessions for the redesign batch | +| Execution runtime | hapi | +| Execution control | HAPI Hub helper spawn/inspect/message; visible sessions in Owner sidebar | +| Execution machine | local: id=a15888ad-d405-4884-a407-c28ddd8b397c; host=wusumacdeMac-mini.local | +| Execution flavor | codex | +| Model selection policy | fixed | +| Fixed model/effort | model=gpt-5.6-sol; effort=xhigh | +| Model catalog evidence | codex-cli 0.146.0; native catalog accepted gpt-5.6-sol with xhigh; per-session JSONL retained | +| Execution permission | yolo | +| Permission scope | runtime-only; Human Gates remain | +| Execution visibility | visible | +| Execution fallback | none; failed HAPI spawn remains inbox and is reported | +| Pre-dispatch recovery | Identity/runtime failure before ACK with clean worktree and zero Goal edits is PMO-owned recovery, not a product attempt; retry automatically while runtime/scope/model/permission/cost remain unchanged | +| Monitoring wait primitive | HAPI completion/blocker event + real timer cell; startup 90s, steady 600s; resume only with functions.wait(real cell_id) | +| Monitoring observe primitive | inspect_peer/session metadata/messages + exact task ledger reconciliation | +| Monitoring target evidence | Region-art R6 HAPI `c63c1b13-8f65-479c-ad65-b2c998713a2d` and final-prototype R6 HAPI `397e65db-795c-46c4-af7a-5eccf23ab56e` pre-dispatch VERIFIED | + +## 共享文件锁 + +PMO 独占 `.agent-taskgraph/`、root README/AGENTS and cross-domain indexes. `ABG-V2-CORE-SYSTEMS` exclusively writes four named files under `docs/product/`; `ABG-V2-CONTENT-MATRIX` exclusively writes `docs/content/`; UX, art and audio workers each own disjoint named files under `docs/presentation/`; the validator owns only `docs/reviews/DESIGN_VALIDATION_CONTRACT.md` and `tools/validate_predevelopment_design.py`. R3 repair reuses disjoint owners: UX writes the two UX authorities, art writes only the art catalog, audio writes only the audio catalog, reference writes only `TRANSFERABLE_STRUCTURE_LESSONS.md`, validator writes only its contract/tool, and Systems writes only `ECONOMY_AND_BALANCE.md` plus the new anti-dominance fixture/script. No worker edits `docs/prototype/` in this batch. + +## 岗位工作区映射 + +| 岗位 | 负责模块/目录 | 常用验收命令 | +|---|---|---| +| reference-researcher | `docs/reference/` | evidence labels and local-link audit | +| systems-designer | `docs/product/` | contract consistency and decision audit | +| content-architect | `docs/content/` | exact-count/cross-reference/no-filler audit | +| presentation-designer | `docs/presentation/`, `docs/prototype/` | state/screen/asset coverage | +| independent-reviewer | `docs/reviews/` | fixed-revision read-only review | + +## 本机环境 + +| 能力 | 状态 | 说明 | +|---|---|---| +| Git/worktree | ✅ | clean main; two redesign worktrees will bind to the PMO baseline commit | +| codex 登录 | ✅ | codex-cli 0.146.0 | +| 开新会话可用路径 | ✅ | HAPI Hub local runner READY; spawn/inspect/message/catalog verified | +| Owner 可见性偏好 | 所有本批 worker/reviewer 均 HAPI 可见 | +| codex 线程工具 | 宿主未暴露 create_thread；使用正式 HAPI Hub spawn，不使用不可见 spawn_agent 冒充会话 | +| 监控机制登记 | R5 dual review active after exact ACKs; startup 90s then steady 600s | + +## 归档约定 + +STATUS 由 queue/ledger 派生；完成任务按月归档，聊天不作为唯一事实源。 diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/README.md b/.agent-taskgraph/archive/recruit-line-redesign-02/README.md new file mode 100644 index 00000000..24a4f34d --- /dev/null +++ b/.agent-taskgraph/archive/recruit-line-redesign-02/README.md @@ -0,0 +1,8 @@ +# Archived Batch: recruit-line-redesign-02 + +This directory preserves the historical TaskGraph control state for the pre-development +redesign batch. It is append-only evidence and is not the current queue, role registry, +status board, spec or graph. + +It was archived because Phase 9 Godot runtime work is active and the historical graph +contains deprecated beta.10 topology that does not validate under beta.12. diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/ROLES.md b/.agent-taskgraph/archive/recruit-line-redesign-02/ROLES.md new file mode 100644 index 00000000..c414a4fb --- /dev/null +++ b/.agent-taskgraph/archive/recruit-line-redesign-02/ROLES.md @@ -0,0 +1,48 @@ +# 角色注册表 + +> Team revision: `rev-13` + +| Role ID | 名称 | 生命周期 | 状态 | 核心职责 | 当前 Goal | Session ID | 最后更新 | +|---|---|---|---|---|---|---|---| +| systems-designer | 系统与数值设计师 | persistent | assigned | 横向长队列、招募、战斗、经济、失败与保存合同 | task:ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR | 22caeb51-307e-41a6-af3e-875b8e37ba6b | 2026-08-11 | +| content-architect | 内容架构师 | persistent | reserved | 生成式成员、职业、特性、装备、敌人、Boss 与区域矩阵 | task:ABG-V2-CONTENT-OUTCOME-SYNC | HAPI 593d50d6 | 2026-08-11 | +| presentation-designer | 横屏体验与信息架构师 | persistent | assigned | 完整页面、状态、引导、输入、可访问性与 UI 语法 | task:ABG-V2-ONBOARDING-PRESENTATION-REPAIR | c5b75ce7-3baa-45aa-a037-c7f8491c68ee | 2026-08-11 | +| art-director-v2 | 角色场景与动效美术总监 | persistent | assigned | 模块化成员、职业、敌人、区域、动画与 VFX 生产规格 | task:ABG-V2-ART-SCHEMA-ORDER-REPAIR | 28cb1721-b66f-4fdf-9700-5ba5924761fa | 2026-08-11 | +| audio-director-v2 | 游戏音频总监 | persistent | assigned | 音乐、环境、SFX、语音、字幕与生命周期音频规格 | task:ABG-V2-AUDIO-COVERAGE-REPAIR | 00ed3a77-7d43-4e45-8dd0-a0e687775b5c | 2026-08-11 | +| design-validator-v2 | 设计验证工程师 | task-scoped | retired | 设计计数、ID、交叉引用、退役概念和覆盖率验证 | task:ABG-V2-CROSSDOC-VALIDATOR-H1 | 0eb4b8b3-4830-4ab1-bfbb-b5a643ce720d | 2026-08-11 | +| reference-researcher | 参考普查分析师 | persistent | assigned | 证据边界、结构经验和当前原创范围的一致性 | task:ABG-V2-REFERENCE-SCOPE-REPAIR | 8e9f4df2-06de-4660-953d-8aa3a9c31199 | 2026-08-11 | +| independent-reviewer-v2 | 独立设计审查员（R1 历史） | task-scoped | retired | 固定 revision 只读检查参考边界与新版一致性 | task:ABG-V2-REVIEW | HAPI 5161887f-fc11-4850-92b7-321ac8aa1bc2 | 2026-08-11 | +| systems-integrity-reviewer-v2 | 系统完整性审查员 | task-scoped | retired | 核心循环、战斗、经济、节奏、存档、失败与内容机械一致性 | task:ABG-V2-SYSTEMS-REVIEW-R2 | HAPI 7945e8ff-abe9-45ef-9f22-2838dacfa197 | 2026-08-11 | +| experience-production-reviewer-v2 | 体验与生产审查员 | task-scoped | retired | 页面可理解性、内容呈现、美术动画、音频、可访问性与生产可行性 | task:ABG-V2-EXPERIENCE-REVIEW-R2 | HAPI 8cfe0437-3a7e-4ac0-8b26-3477ef170654 | 2026-08-11 | +| systems-integrity-reviewer-r3 | 系统完整性独立复审员 R3 | task-scoped | retired | 固定版本核心循环、战斗、经济、节奏、存档、失败与机械内容一致性 | task:ABG-V2-SYSTEMS-REVIEW-R3 | afcafe40-d67b-4372-a63a-6cb94c2e6127 | 2026-08-11 | +| experience-production-reviewer-r3 | 体验与生产独立复审员 R3 | task-scoped | retired | 固定版本可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R3 | 44652407-b548-4f47-b78e-fca47657103f | 2026-08-11 | +| systems-integrity-reviewer-r4 | 系统完整性独立复审员 R4 | task-scoped | retired | 固定版本核心循环、战斗、经济、节奏、存档、失败与机械内容一致性 | task:ABG-V2-SYSTEMS-REVIEW-R4 | ba21ab5a-b78f-4c30-ad12-94b457dcec71 (archived) | 2026-08-11 | +| experience-production-reviewer-r4 | 体验与生产独立复审员 R4 | task-scoped | retired | 固定版本可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R4 | 0e8eec99-d95f-4b98-a9bf-cb5ac46433e1 (archived) | 2026-08-11 | +| run-creation-contract-repairer-r5 | 开局种子与事务合同修复员 | task-scoped | retired | 统一开局 seed、offer receipt、选择恢复与 current_run 原子提交 | task:ABG-V2-RUN-CREATION-TRANSACTION-REPAIR | 258f86c1-239c-4118-b782-55d7ed79c63b (archived) | 2026-08-11 | +| hidden-profession-art-repairer-r5 | 隐藏职业美术绑定修复员 | task-scoped | retired | 对齐 PF-H01..H06 父职业、轮廓、道具、动作与生产规格 | task:ABG-V2-HIDDEN-PROFESSION-ART-REPAIR | c32326b0-6952-46d2-b3b2-31347fa55ee4 (archived) | 2026-08-11 | +| design-validator-r5-h2 | R4 矛盾回归验证工程师 | task-scoped | retired | 为 seed/事务与隐藏职业美术绑定建立语义正负回归门 | task:ABG-V2-R4-VALIDATOR-H2 | 0ab9a014-89c7-4527-857c-6928349d842a (archived) | 2026-08-11 | +| design-validator-r5-h2-a2 | R4 矛盾回归验证工程师（替代尝试） | task-scoped | retired | 为 seed/事务与隐藏职业美术绑定建立语义正负回归门 | task:ABG-V2-R4-VALIDATOR-H2 | 2cf38e48-f8f9-4c61-adff-b822831c17c7 (archived) | 2026-08-11 | +| design-validator-r5-h2-a3 | R4 矛盾回归验证工程师（握手恢复） | task-scoped | retired | 为 seed/事务与隐藏职业美术绑定建立语义正负回归门 | task:ABG-V2-R4-VALIDATOR-H2 | 1546960c-baf0-4a36-bd71-b1c39bc577b7 (archived) | 2026-08-11 | +| systems-integrity-reviewer-r5 | 系统完整性独立复审员 R5 | task-scoped | retired | 固定源核心循环、战斗、经济、节奏、存档、失败与机械内容一致性 | task:ABG-V2-SYSTEMS-REVIEW-R5 | a0ea4f06-a73d-47c4-bc0d-dad36c6d735d (archived) | 2026-08-11 | +| experience-production-reviewer-r5 | 体验与生产独立复审员 R5 | task-scoped | retired | 固定源可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R5 | c52000b5-2021-48cb-8d68-299f36876b8c (archived INVALID) | 2026-08-11 | +| experience-production-reviewer-r5-a2 | 体验与生产独立复审员 R5 替代 | task-scoped | retired | 固定源可理解性、页面覆盖、美术音频生产、可访问性与原创性 | task:ABG-V2-EXPERIENCE-REVIEW-R5 | 2392ec4d-83e8-4df4-b06e-0a7d3fa8167b (archived) | 2026-08-11 | +| region-art-binding-repairer-r6 | 区域内容到美术生产绑定修复员 | task-scoped | assigned | 对齐 RG-01..08 与 ART-REG-01..08，并补语义回归门 | task:ABG-V2-REGION-ART-BINDING-R6 | c63c1b13-8f65-479c-ad65-b2c998713a2d | 2026-08-11 | +| final-prototype-builder-r6 | 最终横屏 HTML 原型制作员 | task-scoped | assigned | 将当前 50 页面、8 区域和完整目录实现为 Owner 可审的 Gallery 原型 | task:ABG-V2-FINAL-PROTOTYPE-R6 | 397e65db-795c-46c4-af7a-5eccf23ab56e | 2026-08-11 | + +## 编制变更历史 + +| 时间 | Team revision | 动作 | Owner 授权 | 结果 | +|---|---|---|---|---| +| 2026-08-10 | rev-1 | 旧固定角色/三路线设计批次 | Owner 初始授权 | Owner 后续停止，历史保留 | +| 2026-08-10 | rev-2 | 两个写入会话并行 + 一个 Owner 手动开启的可见 reviewer | Owner 明确允许多个会话继续 | 等待 worker runtime 验证 | +| 2026-08-10 | rev-3 | 新增 UX、美术、音频、设计验证四个互斥写入会话 | Owner：`可以再多几个会话，10个以内我觉得都OK` | 执行并发扩大到六个 worker；reviewer 后置待命 | +| 2026-08-11 | rev-4 | R1 失败修复后拆为系统完整性与体验生产两个并行只读 reviewer | Owner：`可以再多几个会话，10个以内我觉得都OK` | 六个写入会话已交候选；新增两个 reviewer，总量不超过 10 | +| 2026-08-11 | rev-5 | Experience R3 FAIL 后复用 UX/Art/Audio/Validator 会话并新增一个 Reference 修复会话 | 同一 Owner 并发授权；不扩大产品范围 | 五个互斥修复 Goal；Systems R3 继续只读固定版本 | +| 2026-08-11 | rev-6 | H1 集成全绿后新增两个全新 R4 独立 reviewer | Owner 已授权最多 10 个可见会话与 fresh 双审 | 两个 reviewer 只读同一固定源 `db932cf`；R1-R3 结论只作历史 | +| 2026-08-11 | rev-7 | R4 双 FAIL 后新增两个互斥 task-scoped 修复角色 | 既有 Owner 并发授权；失败路由不扩大产品范围 | seed/transaction 三文档与 hidden-art 单文档并行，之后 validator H2 + fresh R5 | +| 2026-08-11 | rev-8 | R4 两项修复整合后新增 task-scoped H2 validator | graph-r10 已批准失败路由；既有 HAPI profile 授权 | 只改验证合同与脚本；完成后 fresh R5 双审 | +| 2026-08-11 | rev-9 | H2 attempt-1 身份流程无效后替换 task-scoped validator | Owner：`OK的啊继续吧`，明确批准 `1/1 -> 2/2` replacement attempt | 旧角色保持 retired；新角色独占 attempt-2，两文件范围不变 | +| 2026-08-11 | rev-10 | 将零编辑 pre-dispatch 身份失败改为 PMO 自动恢复 | Owner：`你继续啊，为什么一直要我授权` | 新建握手恢复角色；两阶段 dispatch；不消耗产品 Goal 尝试 | +| 2026-08-11 | rev-11 | H2 整合后新增两个 fresh R5 独立 reviewer | 既有 Owner 最多 10 个可见会话及双 reviewer 授权 | 两个只读角色并行审同一固定源；双 PASS 才进入 synthesis | +| 2026-08-11 | rev-12 | Experience R5 首次 harness INVALID 后自动替换 | PROJECT pre-dispatch/reviewer environment recovery policy | 原 reviewer 归档；fresh a2 仍只读同一固定源 | +| 2026-08-11 | rev-13 | Experience R5 Gate 8 后并行区域绑定修复与最终 HTML 制作 | Owner 要求尽快完成完整策划、美术与可查看输出；既有最多 10 个可见会话授权 | 两个写入域互斥；不进入引擎开发或最终媒体生产 | diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/STATUS.md b/.agent-taskgraph/archive/recruit-line-redesign-02/STATUS.md new file mode 100644 index 00000000..a684bfd2 --- /dev/null +++ b/.agent-taskgraph/archive/recruit-line-redesign-02/STATUS.md @@ -0,0 +1,45 @@ +# 任务总览板 + +| 任务 ID | 标题 | 状态 | 负责人 | 轮次 | 最后更新 | 卡点 | +|---|---|---|---|---|---|---| +| ABG-V2-CORE-SYSTEMS | 横向长队列核心系统重构 | review | systems-designer | 2/2 | 2026-08-11T00:47:43+08:00 | repair `e598eed` 已整合为 `9be1927`；等待双 reviewer | +| ABG-V2-CONTENT-MATRIX | 生成式招募与完整内容矩阵 | review | content-architect | 1/2 | 2026-08-10T23:41:16+08:00 | candidates `4e07a91` + `6a41189` 已整合；ACK 到达顺序偏晚已记录 | +| ABG-V2-UX-SCREENS | 横屏 UX 与完整页面状态 | review | presentation-designer | 2/2 | 2026-08-11T00:47:43+08:00 | repair `c64b99e` 已整合为 `ec4863a`；页面 50、绑定 33 | +| ABG-V2-ART-DIRECTION | 模块化角色、场景、动画与 VFX 规格 | review | art-director-v2 | 1/2 | 2026-08-10T23:40:01+08:00 | candidate `ca4d3df` 已整合 | +| ABG-V2-AUDIO-DIRECTION | 音乐、环境、SFX 与语音规格 | review | audio-director-v2 | 2/2 | 2026-08-11T00:47:43+08:00 | repair `ff4101a` 已整合为 `f0095af`；17 个绑定已关闭 | +| ABG-V2-DESIGN-VALIDATION | 完整设计机械验证合同与工具 | review | design-validator-v2 | 2/2 | 2026-08-11T00:47:43+08:00 | repair `e7f80d0` 已整合为 `83bba23`；self-test/full 均通过 | +| ABG-V2-REVIEW | 独立只读完整设计审查 | failed | independent-reviewer-v2 | 1/1 | 2026-08-11T00:34:00+08:00 | FAIL：最后成员解雇可形成未定义 0 人 Market；修复后需新 reviewer | +| ABG-V2-SYSTEMS-REVIEW-R2 | 系统、经济、战斗与存档完整性审查 | failed | systems-integrity-reviewer-v2 | 1/1 | 2026-08-11T01:10:00+08:00 | FAIL Gate 5：PF-A04 依赖未定义 hit/miss/evasion 系统 | +| ABG-V2-EXPERIENCE-REVIEW-R2 | UX、内容呈现与美术音频生产审查 | failed | experience-production-reviewer-v2 | 1/1 | 2026-08-11T01:10:00+08:00 | INVALID：PMO context 引用了固定 revision 中不存在的文件名 | +| ABG-V2-BATTLE-OUTCOME-REPAIR | 确定性战斗结果词汇修复 | review | systems-designer | 1/1 | 2026-08-11T08:25:23+08:00 | candidate `5a05ffc` 整合为 `f2a3f3c`；三项验证通过 | +| ABG-V2-CONTENT-OUTCOME-SYNC | 内容结果词汇同步 | review | content-architect | 1/1 | 2026-08-11T09:14:42+08:00 | `d5fd2fc` -> `edbd281` 已集成冻结；等待 repaired 双审 | +| ABG-V2-ART-OUTCOME-SYNC | 美术与 VFX 结果词汇同步 | review | art-director-v2 | 1/1 | 2026-08-11T09:14:42+08:00 | `8c8c1b7` -> `f8d5bfe` 已集成冻结；新任务只修 schema/order | +| ABG-V2-AUDIO-OUTCOME-SYNC | 音频结果词汇同步 | review | audio-director-v2 | 1/1 | 2026-08-11T09:14:42+08:00 | `ea69bc7` -> `404e996` 已集成冻结；新任务只修 coverage | +| ABG-V2-SYSTEMS-REVIEW-R3 | 系统完整性固定版本复审 | failed | systems-integrity-reviewer-r3 | 1/1 | 2026-08-11T09:36:00+08:00 | FAIL Gate 6：缺五个可执行固定种子反支配 fixture | +| ABG-V2-EXPERIENCE-REVIEW-R3 | 体验与生产固定版本复审 | failed | experience-production-reviewer-r3 | 1/1 | 2026-08-11T09:14:42+08:00 | FAIL：缺 provisional choose-two 页面；另有方向/schema/audio/reference 矛盾 | +| ABG-V2-ONBOARDING-PRESENTATION-REPAIR | 开局选择、引导与队列方向修复 | review | presentation-designer | 1/1 | 2026-08-11T10:04:32+08:00 | `388d90b` -> `14ea46e` 已整合；待 R4 | +| ABG-V2-ART-SCHEMA-ORDER-REPAIR | Recruit 美术 schema 与队列方向修复 | review | art-director-v2 | 1/1 | 2026-08-11T10:04:32+08:00 | `7f8d4b4` -> `03547a7` 已整合；late ACK 偏差保留；待 R4 | +| ABG-V2-AUDIO-COVERAGE-REPAIR | OUT-006 与 50 页面音频覆盖修复 | review | audio-director-v2 | 1/1 | 2026-08-11T10:04:32+08:00 | `9f63e46` -> `7071887` 已整合；late ACK 偏差保留；待 R4 | +| ABG-V2-REFERENCE-SCOPE-REPAIR | 参考经验中的旧固定角色范围修复 | review | reference-researcher | 1/1 | 2026-08-11T10:04:32+08:00 | `565ba2e` -> `96ea49a` 已整合；待 R4 | +| ABG-V2-CROSSDOC-VALIDATOR-REPAIR | R3 跨文档矛盾回归门 | failed | design-validator-v2 | 1/1 | 2026-08-11T10:17:06+08:00 | INVALID：真实整合文档出现 3 个 parser 假阴性；转 H1 | +| ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR | 固定种子构筑反支配模拟修复 | review | systems-designer | 1/1 | 2026-08-11T10:17:06+08:00 | `cb53794` -> `b65646c` 已整合；模拟与 hashes 通过；待 H1/R4 | +| ABG-V2-CROSSDOC-VALIDATOR-H1 | 真实文档 validator 集成修复 | review | design-validator-v2 | 1/1 | 2026-08-11T10:44:55+08:00 | `49e128f` -> `bcb2931` 已整合；20-case/self/full/JSON 与反支配回放全绿；待 R4 双审 | +| ABG-V2-RUN-CREATION-TRANSACTION-REPAIR | 开局 seed 与可恢复事务修复 | review | run-creation-contract-repairer-r5 | 2/2 | 2026-08-11T11:58:55+08:00 | `ad0a57d` -> `bba0c43` 已整合；等待 H2 + fresh Systems R5 | +| ABG-V2-HIDDEN-PROFESSION-ART-REPAIR | 隐藏职业内容到美术绑定修复 | review | hidden-profession-art-repairer-r5 | 2/2 | 2026-08-11T11:58:55+08:00 | `fef9057` -> `8102bd2` 已整合；`c558d19` 禁止；等待 H2 + fresh Experience R5 | +| ABG-V2-R4-VALIDATOR-H2 | R4 seed/事务与隐藏职业绑定回归门 | done | design-validator-r5-h2-a3 | product 1/2; dispatch recovery 3 | 2026-08-11T13:06:00+08:00 | candidate `fb20b66` 整合为 `7dc584d`；24 cases/errors=0/JSON/5 seeds 通过；等待 R5 双审 | +| ABG-V2-SYSTEMS-REVIEW-R5 | 系统完整性固定源复审 | failed | systems-integrity-reviewer-r5 | 1/1 | 2026-08-11T13:25:00+08:00 | reviewer Gate 1-8 PASS；Gate 9 使用错误时长 Goal，PMO 分类 INVALID_REVIEW_CONTRACT；Gate 10 未审 | +| ABG-V2-EXPERIENCE-REVIEW-R5 | 体验与生产固定源复审 | failed | experience-production-reviewer-r5-a2 | product judgment 1/1; environment recovery 2 | 2026-08-11T13:48:00+08:00 | Gates 1-7 PASS；Gate 8 FAIL：ART-REG 未绑定 RG 且序号简报冲突；9-10 未提升 | +| ABG-V2-REGION-ART-BINDING-R6 | 区域美术绑定收口 | active | region-art-binding-repairer-r6 | 1/1 | 2026-08-11T13:57:00+08:00 | exact rev-13 ACK verified；发送完整 Goal | +| ABG-V2-FINAL-PROTOTYPE-R6 | 最终横屏 HTML/Gallery 原型 | active | final-prototype-builder-r6 | 1/1 | 2026-08-11T13:57:00+08:00 | exact rev-13 ACK verified；发送完整 Goal | + +## 批次 + +- `recruit-line-redesign-02` +- Frozen spec: `design-spec-r3` +- Runtime: HAPI visible Codex `gpt-5.6-sol / xhigh / yolo` +- 并发上限：6；H2 已整合；正在冻结同一 R5 source 并准备两个全新只读 reviewer。 +- HTML、最终美术/音频生成、运行时、打包和发布均未授权。 + +## 当前批次总结 + +- R4 两项产品修复及 H2 回归门已整合。下一步只对同一不可变 R5 source 做 Systems/Experience 双审；只有双 PASS 才进入 PMO synthesis 与 Owner 设计门。HTML 与正式开发继续冻结。 diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/graph.yaml b/.agent-taskgraph/archive/recruit-line-redesign-02/graph.yaml new file mode 100644 index 00000000..7ff466ce --- /dev/null +++ b/.agent-taskgraph/archive/recruit-line-redesign-02/graph.yaml @@ -0,0 +1,477 @@ +version: 1 +batch: + id: "recruit-line-redesign-02" + revision: "graph-r11" + spec: ".agent-taskgraph/spec.md" + spec_revision: "design-spec-r3" + approved_by: "Owner" + approved_at: "2026-08-10T22:38:28+08:00" + max_concurrency: 6 + +nodes: + - id: "v2_core_systems" + title: "Recruit-line core systems redesign" + kind: "expert" + goal_ref: "task:ABG-V2-CORE-SYSTEMS" + needs: [] + consumes: [".agent-taskgraph/spec.md", "docs/reference/"] + produces: ["docs/product/GAME_PRODUCT_CONTRACT.md", "docs/product/SYSTEMS_AND_BATTLE.md", "docs/product/ECONOMY_AND_BALANCE.md", "docs/product/SAVE_AND_FAILURE_CONTRACT.md"] + writes: ["docs/product/"] + acceptance: ["Horizontal ordered-line auto-battle replaces lane grid", "Recruitment/death/dismissal/capacity are executable", "Five pacing profiles and save/failure boundaries are consistent"] + on_pass: "v2_integration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_content_matrix" + title: "Generated recruit and complete content matrices" + kind: "expert" + goal_ref: "task:ABG-V2-CONTENT-MATRIX" + needs: [] + consumes: [".agent-taskgraph/spec.md", "docs/reference/"] + produces: ["docs/content/"] + writes: ["docs/content/"] + acceptance: ["42 professions, 36 traits, 320 equipment, 60 artifacts, 100 enemies, 16 bosses and 8 regions", "No fixed protagonist relationships or filler rows", "All IDs and mechanics validate"] + on_pass: "v2_integration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_ux_screens" + title: "Landscape UX and complete screen-state architecture" + kind: "expert" + goal_ref: "task:ABG-V2-UX-SCREENS" + needs: [] + consumes: [".agent-taskgraph/spec.md", "docs/reference/"] + produces: ["docs/presentation/SCREEN_AND_STATE_MAP.md", "docs/presentation/UI_AND_VISUAL_SYSTEM.md"] + writes: ["docs/presentation/SCREEN_AND_STATE_MAP.md", "docs/presentation/UI_AND_VISUAL_SYSTEM.md"] + acceptance: ["All player-facing landscape screens and states are covered", "Retired fixed-cast, lane-grid and manual-command UI is absent", "Shop, recruit, equip, reorder, auto-battle and recovery are immediately readable"] + on_pass: "v2_integration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_art_direction" + title: "Generated-recruit art, animation and VFX production bible" + kind: "expert" + goal_ref: "task:ABG-V2-ART-DIRECTION" + needs: [] + consumes: [".agent-taskgraph/spec.md", "docs/reference/"] + produces: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] + writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] + acceptance: ["Modular recruits and 42 profession identities are production-countable", "Eight regions and 116 enemy identities have readable art/animation coverage", "Horizontal-line combat remains legible at target landscape sizes"] + on_pass: "v2_integration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_audio_direction" + title: "Music, ambience, SFX and voice production bible" + kind: "expert" + goal_ref: "task:ABG-V2-AUDIO-DIRECTION" + needs: [] + consumes: [".agent-taskgraph/spec.md", "docs/reference/"] + produces: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] + writes: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] + acceptance: ["Audio makes shop, order, battle and failure consequences readable", "Voice scope matches generated recruits and no fixed-cast burden remains", "Production counts, priority, accessibility and lifecycle behavior are explicit"] + on_pass: "v2_integration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_design_validation" + title: "Pre-development design validator and coverage contract" + kind: "expert" + goal_ref: "task:ABG-V2-DESIGN-VALIDATION" + needs: [] + consumes: [".agent-taskgraph/spec.md", "docs/reference/"] + produces: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] + writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] + acceptance: ["Expected counts and stable IDs are mechanically checkable", "Retired concepts and cross-reference failures are detected", "Validator reports unknown pending bindings without inventing product authority"] + on_pass: "v2_integration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_integration" + title: "Integrate and reconcile the redesign" + kind: "merge" + goal_ref: "task:ABG-V2-INTEGRATION" + needs: ["v2_core_systems", "v2_content_matrix", "v2_ux_screens", "v2_art_direction", "v2_audio_direction", "v2_design_validation"] + consumes: ["docs/product/", "docs/content/", "docs/presentation/", "docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] + produces: ["clean main revision", "contract validation report"] + writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] + acceptance: ["No cross-document contradiction", "Repository validator passes", "All six design authorities are integrated", "Fixed revision ready for independent review"] + on_pass: "v2_systems_review" + on_fail: "failed" + max_attempts: 1 + - id: "v2_systems_review" + title: "Independent systems integrity review" + kind: "verifier" + goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R2" + needs: ["v2_integration"] + consumes: ["fixed integrated revision"] + produces: ["read-only systems PASS/FAIL/INVALID review"] + writes: [] + acceptance: ["Core loop, battle, economy, pacing, save/failure and mechanical content coherence pass"] + on_pass: "v2_review_synthesis" + on_fail: "v2_battle_outcome_repair" + max_attempts: 1 + + - id: "v2_experience_review" + title: "Independent experience and production review" + kind: "verifier" + goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R2" + needs: ["v2_integration"] + consumes: ["fixed integrated revision"] + produces: ["read-only experience PASS/FAIL/INVALID review"] + writes: [] + acceptance: ["UX comprehension, content presentation, art/animation/audio feasibility, accessibility and originality pass"] + on_pass: "v2_review_synthesis" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_battle_outcome_repair" + title: "Define deterministic battle outcome vocabulary" + kind: "expert" + goal_ref: "task:ABG-V2-BATTLE-OUTCOME-REPAIR" + needs: ["v2_systems_review"] + consumes: ["systems R2 FAIL", "docs/product/SYSTEMS_AND_BATTLE.md", "all current miss/dodge/evasion references"] + produces: ["one executable deterministic outcome contract"] + writes: ["docs/product/SYSTEMS_AND_BATTLE.md"] + acceptance: ["No accuracy/evasion/hidden-miss probability is introduced", "Every applied/invalidated/resisted/immune/intercepted outcome is executable and journaled", "Skill schema can bind deterministic outcome triggers"] + on_pass: "v2_content_outcome_sync" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_content_outcome_sync" + title: "Synchronize profession and content outcome semantics" + kind: "expert" + goal_ref: "task:ABG-V2-CONTENT-OUTCOME-SYNC" + needs: ["v2_battle_outcome_repair"] + consumes: ["deterministic outcome contract"] + produces: ["content rows bound to declared outcomes"] + writes: ["docs/content/PROFESSION_CATALOG.md", "docs/content/EQUIPMENT_CATALOG.md", "docs/content/TRAIT_CATALOG.md", "docs/content/ENEMY_CATALOG.md"] + acceptance: ["PF-A04 is executable without accuracy/evasion", "All miss/dodge wording maps to a declared outcome or is rewritten", "No content count or unrelated row changes"] + on_pass: "v2_outcome_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_art_outcome_sync" + title: "Synchronize art and VFX outcome semantics" + kind: "expert" + goal_ref: "task:ABG-V2-ART-OUTCOME-SYNC" + needs: ["v2_battle_outcome_repair"] + consumes: ["deterministic outcome contract"] + produces: ["art/VFX cues bound to declared outcomes"] + writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] + acceptance: ["Avoidance cues name only declared deterministic outcomes", "Readability and reduced-motion variants remain explicit"] + on_pass: "v2_outcome_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_audio_outcome_sync" + title: "Synchronize audio outcome semantics" + kind: "expert" + goal_ref: "task:ABG-V2-AUDIO-OUTCOME-SYNC" + needs: ["v2_battle_outcome_repair"] + consumes: ["deterministic outcome contract"] + produces: ["SFX/caption cues bound to declared outcomes"] + writes: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] + acceptance: ["Dodge/miss cues name only declared deterministic outcomes", "Bindings, production counts and accessibility remain valid"] + on_pass: "v2_outcome_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_outcome_reintegration" + title: "Reintegrate deterministic battle outcome repair" + kind: "merge" + goal_ref: "task:ABG-V2-OUTCOME-REINTEGRATION" + needs: ["v2_battle_outcome_repair", "v2_content_outcome_sync", "v2_art_outcome_sync", "v2_audio_outcome_sync"] + consumes: ["repaired product, content, art and audio authorities"] + produces: ["new fixed review revision", "validator and semantic reference audit"] + writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] + acceptance: ["Validator remains green", "No undeclared accuracy/evasion/miss mechanic remains", "R2 FAIL and INVALID are append-only history"] + on_pass: "v2_systems_review_r3" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_systems_review_r3" + title: "Fresh systems integrity re-review" + kind: "verifier" + goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R3" + needs: ["v2_outcome_reintegration"] + consumes: ["new fixed integrated revision"] + produces: ["read-only systems PASS/FAIL/INVALID terminal"] + writes: [] + acceptance: ["All systems gates rerun from zero, including deterministic outcome vocabulary"] + on_pass: "v2_review_synthesis" + on_fail: "v2_anti_dominance_simulation_repair" + max_attempts: 1 + + - id: "v2_anti_dominance_simulation_repair" + title: "Add executable fixed-seed anti-dominance fixtures" + kind: "expert" + goal_ref: "task:ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR" + needs: ["v2_systems_review_r3"] + consumes: ["Systems R3 Gate 6 FAIL", "economy/build policy authority"] + produces: ["five exact seed fixtures", "executable policy comparison", "documented counterfactual outcomes"] + writes: ["docs/product/ECONOMY_AND_BALANCE.md", "docs/product/fixtures/ANTI_DOMINANCE_FIXED_SEEDS.json", "tools/simulate_anti_dominance.py"] + acceptance: ["Five fixed seeds and all inputs are reproducible", "No policy dominates all legal states", "Self-test rejects a deliberately dominant fixture"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_systems_review_r4" + title: "Fresh systems integrity re-review after anti-dominance repair" + kind: "verifier" + goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R4" + needs: ["v2_experience_reintegration"] + consumes: ["new immutable repaired revision"] + produces: ["read-only systems PASS/FAIL/INVALID terminal"] + writes: [] + acceptance: ["All systems gates rerun from zero including fixed-seed anti-dominance fixtures"] + on_pass: "v2_review_synthesis" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_experience_review_r3" + title: "Fresh experience and production re-review" + kind: "verifier" + goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R3" + needs: ["v2_outcome_reintegration"] + consumes: ["new fixed integrated revision", "actual reference authority paths"] + produces: ["read-only experience PASS/FAIL/INVALID terminal"] + writes: [] + acceptance: ["All experience gates rerun from zero with valid context paths"] + on_pass: "v2_review_synthesis" + on_fail: "v2_experience_repair_fanout" + max_attempts: 1 + + - id: "v2_experience_repair_fanout" + title: "Repair R3 experience cross-document failures" + kind: "fork" + goal_ref: "decision:ABG-V2-EXPERIENCE-R3-REPAIR" + needs: ["v2_experience_review_r3"] + consumes: ["R3 experience FAIL terminal"] + produces: ["five disjoint repair candidates"] + writes: [] + acceptance: ["Every reported contradiction has one owner and non-overlapping write scope"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_onboarding_presentation_repair" + title: "Repair run creation, onboarding and party-line presentation" + kind: "expert" + goal_ref: "task:ABG-V2-ONBOARDING-PRESENTATION-REPAIR" + needs: ["v2_experience_review_r3"] + consumes: ["product run-creation authority", "R3 experience FAIL"] + produces: ["complete provisional-recruit decision state and one party-line order"] + writes: ["docs/presentation/SCREEN_AND_STATE_MAP.md", "docs/presentation/UI_AND_VISUAL_SYSTEM.md"] + acceptance: ["Four provisional offers and choose-two revision are explicit before run creation", "First exposed position has one cross-document direction", "Screen count remains intentional and validator-readable"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_art_schema_order_repair" + title: "Repair recruit visual schema mapping and party-line order" + kind: "expert" + goal_ref: "task:ABG-V2-ART-SCHEMA-ORDER-REPAIR" + needs: ["v2_experience_review_r3"] + consumes: ["recruit generation schema", "UI party-line authority", "R3 experience FAIL"] + produces: ["exact content-to-art recipe crosswalk and coherent counts"] + writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] + acceptance: ["Every recruit content field maps to an art field or explicit derived assembly", "Counts and totals recalculate", "Party line matches UI authority"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_audio_coverage_repair" + title: "Repair screen and abandonment audio coverage" + kind: "expert" + goal_ref: "task:ABG-V2-AUDIO-COVERAGE-REPAIR" + needs: ["v2_experience_review_r3"] + consumes: ["50-screen authority", "OUT-006 abandonment state", "R3 experience FAIL"] + produces: ["complete screen/action audio binding"] + writes: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] + acceptance: ["OUT-006 is explicitly bound", "Screen total is 50", "Audio production totals remain internally consistent"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_reference_scope_repair" + title: "Repair transferred-lesson scope to generated recruits" + kind: "expert" + goal_ref: "task:ABG-V2-REFERENCE-SCOPE-REPAIR" + needs: ["v2_experience_review_r3"] + consumes: ["design-spec-r3", "current product/content authorities", "R3 experience FAIL"] + produces: ["current original-scope lessons without fixed-cast residue"] + writes: ["docs/reference/TRANSFERABLE_STRUCTURE_LESSONS.md"] + acceptance: ["No named-recruit or relationship requirement remains", "Current generated-recruit and profession scope is exact", "Reference facts remain separate from original design commitments"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_crossdoc_validator_repair" + title: "Add regression gates for R3 experience contradictions" + kind: "expert" + goal_ref: "task:ABG-V2-CROSSDOC-VALIDATOR-REPAIR" + needs: ["v2_experience_review_r3"] + consumes: ["five R3 contradiction classes", "current design validator"] + produces: ["self-tested cross-document regression gates"] + writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] + acceptance: ["Each R3 contradiction class has a positive and negative check", "Existing exact counts remain enforced", "Self-test and full validation pass only after all repair candidates integrate"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_experience_reintegration" + title: "Integrate and freeze R3 experience repairs" + kind: "merge" + goal_ref: "task:ABG-V2-EXPERIENCE-REINTEGRATION" + needs: ["v2_onboarding_presentation_repair", "v2_art_schema_order_repair", "v2_audio_coverage_repair", "v2_reference_scope_repair", "v2_crossdoc_validator_repair", "v2_anti_dominance_simulation_repair"] + consumes: ["five repair candidates", "systems R3 terminal"] + produces: ["new immutable review revision"] + writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] + acceptance: ["All R3 experience blockers are closed", "Systems terminal is reconciled", "Validator self-test/full/JSON pass"] + on_pass: "v2_experience_review_r4" + on_fail: "v2_crossdoc_validator_h1" + max_attempts: 1 + + - id: "v2_crossdoc_validator_h1" + title: "Repair validator integration against real repaired authorities" + kind: "expert" + goal_ref: "task:ABG-V2-CROSSDOC-VALIDATOR-H1" + needs: ["v2_experience_reintegration"] + consumes: ["integrated main false-negative diagnostics", "real repaired UX and art authorities"] + produces: ["real-document parser repair", "positive and negative regression fixtures"] + writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] + acceptance: ["Integrated repaired main passes text and JSON", "20-case self-test remains strict", "Each R3 contradiction mutation still fails independently"] + on_pass: "v2_experience_reintegration" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_experience_review_r4" + title: "Fresh experience and production re-review after R3 repair" + kind: "verifier" + goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R4" + needs: ["v2_experience_reintegration"] + consumes: ["new immutable repaired revision"] + produces: ["read-only experience PASS/FAIL/INVALID terminal"] + writes: [] + acceptance: ["All experience gates rerun from zero; no R3 evidence reused as PASS"] + on_pass: "v2_review_synthesis" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_review_synthesis" + title: "PMO dual-review synthesis" + kind: "merge" + goal_ref: "task:ABG-V2-REVIEW-SYNTHESIS" + needs: ["v2_systems_review_r4", "v2_experience_review_r4"] + consumes: ["systems review terminal", "experience review terminal"] + produces: ["one PMO acceptance or earliest repair route"] + writes: [".agent-taskgraph/queue/review/", ".agent-taskgraph/STATUS.md"] + acceptance: ["Both reviewers used the same fixed revision", "No open reviewer failure or invalid environment", "Open human and implementation gates remain explicit"] + on_pass: "owner_v2_gate" + on_fail: "failed" + max_attempts: 1 + + - id: "owner_v2_gate" + title: "Owner approves complete design before HTML prototype redesign" + kind: "human_gate" + goal_ref: "decision:ABG-V2-OWNER-GATE" + needs: ["v2_review_synthesis_r5"] + consumes: ["integrated design", "independent review"] + produces: ["owner accept or revise decision"] + writes: [] + acceptance: ["Owner authorizes landscape HTML prototype redesign"] + on_pass: "done" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_r4_seed_transaction_repair" + title: "Repair provisional Recruit seed and transaction authority" + kind: "expert" + goal_ref: "task:ABG-V2-RUN-CREATION-TRANSACTION-REPAIR" + needs: ["v2_systems_review_r4"] + consumes: ["Systems R4 Gate 3 FAIL", "run creation, save and recruit generation authorities"] + produces: ["durable pre-run seed allocation", "reproducible offer receipt", "atomic current_run commit and retry rules"] + writes: ["docs/product/GAME_PRODUCT_CONTRACT.md", "docs/product/SAVE_AND_FAILURE_CONTRACT.md", "docs/content/RECRUIT_GENERATION.md"] + acceptance: ["The seed exists durably before offers", "Crash and retry reproduce identical offers/selection", "Run creation commits once without free reroll or duplicate mutation"] + on_pass: "v2_r4_reintegration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_r4_hidden_profession_art_repair" + title: "Repair hidden profession content-to-art bindings" + kind: "expert" + goal_ref: "task:ABG-V2-HIDDEN-PROFESSION-ART-REPAIR" + needs: ["v2_experience_review_r4"] + consumes: ["Experience R4 Gate 7 FAIL", "PF-H01..H06 content production briefs"] + produces: ["exact hidden-profession parent, silhouette, prop and motion bindings"] + writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] + acceptance: ["PF-H01..H06 match content mechanics and production briefs", "No unresolved parent binding remains", "Art totals and other profession rows stay unchanged"] + on_pass: "v2_r4_reintegration" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_r4_reintegration" + title: "Integrate R4 bounded repairs" + kind: "merge" + goal_ref: "task:ABG-V2-R4-REINTEGRATION" + needs: ["v2_r4_seed_transaction_repair", "v2_r4_hidden_profession_art_repair"] + consumes: ["two disjoint repair candidates"] + produces: ["one integrated R4-repaired revision"] + writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] + acceptance: ["Both exact failure boundaries close", "Existing validators and anti-dominance replay remain green"] + on_pass: "v2_r4_validator_h2" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_r4_validator_h2" + title: "Add R4 contradiction regression gates" + kind: "expert" + goal_ref: "task:ABG-V2-R4-VALIDATOR-H2" + needs: ["v2_r4_reintegration"] + consumes: ["integrated repaired authorities", "both R4 FAIL reports"] + produces: ["positive parser coverage and isolated negative mutations for both contradictions"] + writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] + acceptance: ["Real repaired documents pass", "Seed-order/transaction mutation fails", "Hidden-profession art mismatch/unresolved mutation fails"] + on_pass: "v2_systems_review_r5" + on_fail: "failed" + max_attempts: 2 + + - id: "v2_systems_review_r5" + title: "Fresh systems integrity review after R4 repairs" + kind: "verifier" + goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R5" + needs: ["v2_r4_validator_h2"] + consumes: ["new immutable R5 review source"] + produces: ["read-only systems PASS/FAIL/INVALID terminal"] + writes: [] + acceptance: ["All systems gates rerun from zero including provisional seed transaction"] + on_pass: "v2_experience_review_r5" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_experience_review_r5" + title: "Fresh experience and production review after R4 repairs" + kind: "verifier" + goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R5" + needs: ["v2_r4_validator_h2"] + consumes: ["new immutable R5 review source"] + produces: ["read-only experience PASS/FAIL/INVALID terminal"] + writes: [] + acceptance: ["All experience gates rerun from zero including hidden-profession production binding"] + on_pass: "v2_review_synthesis_r5" + on_fail: "failed" + max_attempts: 1 + + - id: "v2_review_synthesis_r5" + title: "PMO R5 dual-review synthesis" + kind: "merge" + goal_ref: "task:ABG-V2-REVIEW-SYNTHESIS-R5" + needs: ["v2_systems_review_r5", "v2_experience_review_r5"] + consumes: ["systems R5 terminal", "experience R5 terminal"] + produces: ["one PMO acceptance or earliest repair route"] + writes: [".agent-taskgraph/queue/review/", ".agent-taskgraph/STATUS.md"] + acceptance: ["Both reviewers used the same fixed revision", "Both terminals PASS", "Human and implementation gates remain explicit"] + on_pass: "owner_v2_gate" + on_fail: "failed" + max_attempts: 1 diff --git a/.agent-taskgraph/manifests/ABG-V2-ART-DIRECTION.json b/.agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-ART-DIRECTION.json similarity index 100% rename from .agent-taskgraph/manifests/ABG-V2-ART-DIRECTION.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-ART-DIRECTION.json diff --git a/.agent-taskgraph/manifests/ABG-V2-AUDIO-DIRECTION.json b/.agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-AUDIO-DIRECTION.json similarity index 100% rename from .agent-taskgraph/manifests/ABG-V2-AUDIO-DIRECTION.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-AUDIO-DIRECTION.json diff --git a/.agent-taskgraph/manifests/ABG-V2-DESIGN-VALIDATION.json b/.agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-DESIGN-VALIDATION.json similarity index 100% rename from .agent-taskgraph/manifests/ABG-V2-DESIGN-VALIDATION.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-DESIGN-VALIDATION.json diff --git a/.agent-taskgraph/manifests/ABG-V2-UX-SCREENS.json b/.agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-UX-SCREENS.json similarity index 100% rename from .agent-taskgraph/manifests/ABG-V2-UX-SCREENS.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/manifests/ABG-V2-UX-SCREENS.json diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/.gitkeep b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/.gitkeep new file mode 100644 index 00000000..e69de29b diff --git a/.agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/context.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/context.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/dispatch.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/goal.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/goal.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/ledger.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/ledger.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-FINAL-PROTOTYPE-R6/runtime-evidence.json diff --git a/.agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/context.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/context.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/dispatch.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/goal.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/goal.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/ledger.md similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/ledger.md diff --git a/.agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/active/ABG-V2-REGION-ART-BINDING-R6/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/active/ABG-V2-REGION-ART-BINDING-R6/runtime-evidence.json diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/.gitkeep b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/.gitkeep new file mode 100644 index 00000000..e69de29b diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt2-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt2-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt2-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/context-attempt2-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/context.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/context.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt2-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt2-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt2-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch-attempt2-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/dispatch.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt2-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt2-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt2-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/goal-attempt2-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/goal.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/goal.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/ledger.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/ledger.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt2-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt2-invalid.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt2-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/report-attempt2-invalid.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/report.md similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/report.md diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt1-invalid.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt1-invalid.json similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt1-invalid.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt1-invalid.json diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt2-invalid.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt2-invalid.json similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt2-invalid.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence-attempt2-invalid.json diff --git a/.agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/done/ABG-V2-R4-VALIDATOR-H2/runtime-evidence.json diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/.gitkeep b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/.gitkeep new file mode 100644 index 00000000..e69de29b diff --git a/.agent-taskgraph/queue/failed/ABG-REF-CENSUS/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-REF-CENSUS/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-REF-CENSUS/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-REF-CENSUS/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-REF-CENSUS/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-REF-CENSUS/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-REF-CENSUS/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-REF-CENSUS/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-REF-CENSUS/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/reviewer-dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/reviewer-dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/reviewer-dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/reviewer-dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/reviewer-runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/reviewer-runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-DESIGN/reviewer-runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-DESIGN/reviewer-runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/dispatch-attempt-1.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/dispatch-attempt-1.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/dispatch-attempt-1.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/dispatch-attempt-1.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/task-manifest.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/task-manifest.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-SYSTEM-REPAIR/task-manifest.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-SYSTEM-REPAIR/task-manifest.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-CROSSDOC-VALIDATOR-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R2/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R3/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R4/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report-attempt1-invalid.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report-attempt1-invalid.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report-attempt1-invalid.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report-attempt1-invalid.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence-attempt1-invalid.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence-attempt1-invalid.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence-attempt1-invalid.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence-attempt1-invalid.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-EXPERIENCE-REVIEW-R5/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-REVIEW/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-REVIEW/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-REVIEW/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-REVIEW/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-REVIEW/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-REVIEW/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-REVIEW/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-REVIEW/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-REVIEW/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-REVIEW/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-REVIEW/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-REVIEW/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-REVIEW/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R2/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R3/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R4/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/ledger.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/report.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/report.md diff --git a/.agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-V2-SYSTEMS-REVIEW-R5/runtime-evidence.json diff --git a/.agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/context.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/context.md diff --git a/.agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/dispatch.md diff --git a/.agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/goal.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/goal.md diff --git a/.agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/ledger.md similarity index 100% rename from .agent-taskgraph/queue/failed/ABG-WORLD-CONTENT/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/failed/ABG-WORLD-CONTENT/ledger.md diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/queue/inbox/.gitkeep b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/inbox/.gitkeep new file mode 100644 index 00000000..e69de29b diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/.gitkeep b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/.gitkeep new file mode 100644 index 00000000..e69de29b diff --git a/.agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-DIRECTION/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-DIRECTION/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-OUTCOME-SYNC/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-OUTCOME-SYNC/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ART-SCHEMA-ORDER-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-COVERAGE-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-DIRECTION/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-DIRECTION/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-AUDIO-OUTCOME-SYNC/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-BATTLE-OUTCOME-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-MATRIX/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-MATRIX/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CONTENT-OUTCOME-SYNC/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CORE-SYSTEMS/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CORE-SYSTEMS/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/report.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/report.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-CROSSDOC-VALIDATOR-H1/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-DESIGN-VALIDATION/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-DESIGN-VALIDATION/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/report.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/report.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence-attempt1-invalid.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence-attempt1-invalid.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence-attempt1-invalid.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence-attempt1-invalid.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-HIDDEN-PROFESSION-ART-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-ONBOARDING-PRESENTATION-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-REFERENCE-SCOPE-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/report.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/report.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/report.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/report.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence-attempt1-invalid.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence-attempt1-invalid.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence-attempt1-invalid.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence-attempt1-invalid.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-RUN-CREATION-TRANSACTION-REPAIR/runtime-evidence.json diff --git a/.agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/context.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/context.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/context.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/context.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/dispatch.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/dispatch.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/dispatch.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/dispatch.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/goal.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/goal.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/goal.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/goal.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/ledger.md b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/ledger.md similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/ledger.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/ledger.md diff --git a/.agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/runtime-evidence.json b/.agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/runtime-evidence.json similarity index 100% rename from .agent-taskgraph/queue/review/ABG-V2-UX-SCREENS/runtime-evidence.json rename to .agent-taskgraph/archive/recruit-line-redesign-02/queue/review/ABG-V2-UX-SCREENS/runtime-evidence.json diff --git a/.agent-taskgraph/roles/art-director-v2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/art-director-v2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/art-director-v2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/art-director-v2/ROLE.md diff --git a/.agent-taskgraph/roles/audio-director-v2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/audio-director-v2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/audio-director-v2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/audio-director-v2/ROLE.md diff --git a/.agent-taskgraph/roles/content-architect/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/content-architect/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/content-architect/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/content-architect/ROLE.md diff --git a/.agent-taskgraph/roles/design-validator-r5-h2-a2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-r5-h2-a2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/design-validator-r5-h2-a2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-r5-h2-a2/ROLE.md diff --git a/.agent-taskgraph/roles/design-validator-r5-h2-a3/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-r5-h2-a3/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/design-validator-r5-h2-a3/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-r5-h2-a3/ROLE.md diff --git a/.agent-taskgraph/roles/design-validator-r5-h2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-r5-h2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/design-validator-r5-h2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-r5-h2/ROLE.md diff --git a/.agent-taskgraph/roles/design-validator-v2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-v2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/design-validator-v2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/design-validator-v2/ROLE.md diff --git a/.agent-taskgraph/roles/experience-production-reviewer-r3/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r3/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/experience-production-reviewer-r3/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r3/ROLE.md diff --git a/.agent-taskgraph/roles/experience-production-reviewer-r4/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r4/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/experience-production-reviewer-r4/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r4/ROLE.md diff --git a/.agent-taskgraph/roles/experience-production-reviewer-r5-a2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r5-a2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/experience-production-reviewer-r5-a2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r5-a2/ROLE.md diff --git a/.agent-taskgraph/roles/experience-production-reviewer-r5/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r5/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/experience-production-reviewer-r5/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-r5/ROLE.md diff --git a/.agent-taskgraph/roles/experience-production-reviewer-v2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-v2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/experience-production-reviewer-v2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/experience-production-reviewer-v2/ROLE.md diff --git a/.agent-taskgraph/roles/final-prototype-builder-r6/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/final-prototype-builder-r6/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/final-prototype-builder-r6/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/final-prototype-builder-r6/ROLE.md diff --git a/.agent-taskgraph/roles/hidden-profession-art-repairer-r5/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/hidden-profession-art-repairer-r5/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/hidden-profession-art-repairer-r5/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/hidden-profession-art-repairer-r5/ROLE.md diff --git a/.agent-taskgraph/roles/independent-reviewer/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/independent-reviewer/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/independent-reviewer/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/independent-reviewer/ROLE.md diff --git a/.agent-taskgraph/roles/presentation-designer/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/presentation-designer/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/presentation-designer/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/presentation-designer/ROLE.md diff --git a/.agent-taskgraph/roles/reference-researcher/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/reference-researcher/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/reference-researcher/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/reference-researcher/ROLE.md diff --git a/.agent-taskgraph/roles/region-art-binding-repairer-r6/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/region-art-binding-repairer-r6/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/region-art-binding-repairer-r6/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/region-art-binding-repairer-r6/ROLE.md diff --git a/.agent-taskgraph/roles/run-creation-contract-repairer-r5/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/run-creation-contract-repairer-r5/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/run-creation-contract-repairer-r5/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/run-creation-contract-repairer-r5/ROLE.md diff --git a/.agent-taskgraph/roles/systems-designer/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-designer/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/systems-designer/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-designer/ROLE.md diff --git a/.agent-taskgraph/roles/systems-integrity-reviewer-r3/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-r3/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/systems-integrity-reviewer-r3/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-r3/ROLE.md diff --git a/.agent-taskgraph/roles/systems-integrity-reviewer-r4/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-r4/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/systems-integrity-reviewer-r4/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-r4/ROLE.md diff --git a/.agent-taskgraph/roles/systems-integrity-reviewer-r5/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-r5/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/systems-integrity-reviewer-r5/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-r5/ROLE.md diff --git a/.agent-taskgraph/roles/systems-integrity-reviewer-v2/ROLE.md b/.agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-v2/ROLE.md similarity index 100% rename from .agent-taskgraph/roles/systems-integrity-reviewer-v2/ROLE.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/roles/systems-integrity-reviewer-v2/ROLE.md diff --git a/.agent-taskgraph/archive/recruit-line-redesign-02/spec.md b/.agent-taskgraph/archive/recruit-line-redesign-02/spec.md new file mode 100644 index 00000000..db591b22 --- /dev/null +++ b/.agent-taskgraph/archive/recruit-line-redesign-02/spec.md @@ -0,0 +1,150 @@ +# Execution Spec: Recruit-Line Auto-Battler Redesign + +> Status: FROZEN +> Requested: 2026-08-10 +> Frozen by: Owner / 2026-08-10 +> Revision: design-spec-r3 + +## Decision This Batch Enables + +Produce a coherent replacement for the current fixed-cast, six-unit, +three-lane design and, in parallel, replace the obsolete presentation +authorities with implementation-ready UX, art, audio and validation contracts. +This batch still does not start runtime development or final media production. + +## Confirmed Product Direction + +- Original landscape 2D Japanese hand-drawn fantasy auto-battler. +- The transferable loop is: shop offers -> recruit/buy -> equip -> reorder one + horizontal party line -> choose risk -> fully automatic battle -> reward, + death, dismissal or replacement -> repeat -> persistent run progression. +- Recruits are compositional units built from generated name, modular visual + identity, fixed profession, rolled attributes, one lightweight mechanical + trait, equipment, level and survival history. +- Recruits are not 30 fixed protagonists. Remove relationship, affinity, + romance, personal-arc and mandatory individual voice systems. +- Battle uses one ordered horizontal line. Remove the 3-lane x front/rear grid, + in-battle Directive system, Anchor/Cache battle terminology and fixed six-unit + cap from the core loop. +- Battle is automated. Player controls are prepare, reorder, equip, start, + pause, speed, inspect and retreat; no repeated manual skill activation. +- Simultaneous deployment starts small and grows toward a simulated target of + 10-12. Cumulative recruits in a run are not capped by that deployment number; + death, dismissal and replacement keep recruitment relevant. + +## Complete Content Envelope + +- 12 base professions, 24 regular advanced professions and 6 hidden hybrid + professions: 42 playable profession identities. +- 36 lightweight mechanical recruit traits. +- 320 equipment items and 60 artifacts. +- 100 ordinary/elite enemy identities plus 16 bosses across 8 regions. +- First clear target 15-25 hours; mastery and high-difficulty play 60+ hours. +- Minimal world framing and regional progression only. Content depth comes from + profession, equipment, formation-order, enemy and run-economy interaction. + +## Workstreams + +### ABG-V2-CORE-SYSTEMS + +Owns root `AGENTS.md` alignment and `docs/product/`. It defines the complete +loop, recruitment/dismissal/death rules, horizontal formation grammar, +deterministic auto-battle, professions as a fixed catalog schema, economy, +shop refresh, pacing, run structure, difficulty, failure, save and postgame. +It must not enumerate the content catalog owned by the content workstream. + +### ABG-V2-CONTENT-MATRIX + +Owns `docs/content/`. It replaces the fixed-character bible with generated +recruit specifications and authors the complete profession, trait, equipment, +artifact, enemy, boss and region matrices. It must consume the schemas in this +spec and must not edit product, presentation, prototype or taskgraph files. + +### ABG-V2-UX-SCREENS + +Owns `docs/presentation/SCREEN_AND_STATE_MAP.md` and +`docs/presentation/UI_AND_VISUAL_SYSTEM.md`. It defines the complete landscape +navigation, shop, recruit, equipment, party-line, battle, result, failure, +settings, accessibility, localization and endgame states. Exact system values +that are still being finalized must use named binding placeholders rather than +inventing a conflicting authority. + +### ABG-V2-ART-DIRECTION + +Owns `docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md`. It replaces fixed +character, relationship and lane-grid assets with a modular generated-recruit +system, 42 profession kits, one horizontal battle line, eight region families, +116 enemy identities, equipment readability, animation states and VFX rules. +It specifies production; it does not generate final assets. + +### ABG-V2-AUDIO-DIRECTION + +Owns `docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md`. It removes the +obsolete 30-character voice burden and specifies readable shop, formation, +auto-battle, casualty, reward, region, UI, accessibility and lifecycle audio. +It specifies production; it does not generate or purchase audio. + +### ABG-V2-DESIGN-VALIDATION + +Owns `docs/reviews/DESIGN_VALIDATION_CONTRACT.md` and +`tools/validate_predevelopment_design.py`. It defines deterministic count, +cross-reference, forbidden-concept, schema and coverage checks for the merged +design. It may not change product, content or presentation authority. + +## Shared Vocabulary + +| Term | Meaning | +|---|---| +| Recruit | One generated run participant, not an authored story protagonist | +| Profession | A fixed reusable combat ruleset referenced by many recruits | +| Trait | One short mechanical modifier with a visible benefit and cost | +| Party line | The ordered simultaneous deployment from front to rear | +| Run | One shop/battle/reward sequence with cumulative recruits and losses | +| Meta progression | Persistent unlock breadth; it may not replace run decisions | +| Artifact | A run-defining shared rule modifier, distinct from unit equipment | + +## Acceptance Criteria + +- No product authority still requires fixed named protagonists, relationships, + a six-unit roster, a three-lane grid or Directive/Anchor/Cache battle rules. +- Recruitment, dismissal, permanent loss, replacement, price pressure, + deployment capacity and cumulative recruitment are explicit and testable. +- Party order changes targeting, protection, reach, healing, movement and + replacement behavior in immediately readable ways. +- Every profession and content row changes a player decision; no name-only or + number-only filler. +- Economy and pacing include five profiles, decision timestamps, solvency and + anti-dominance checks. +- Save/failure rules cover migration, corruption, retry, duplicate claim, + interruption, rollback, clock rollback and bounded offline behavior. +- Cross-document terminology, IDs, counts and formulas are consistent. +- Every player-facing page and key state has a landscape composition, primary + action, readable consequence, empty/error/confirmation behavior and input, + localization and accessibility coverage. +- Art, animation, VFX, music, ambience, SFX and voice scope are production- + countable and match generated recruits rather than fixed protagonists. +- A deterministic design validator checks counts, IDs, cross-references, + forbidden retired concepts and presentation coverage after integration. +- A visible independent reviewer audits the integrated fixed revision before + any HTML prototype or runtime development begins. + +## Excluded Scope + +- Runtime code, engine selection, packages, signing, store or release work. +- Final or bulk production art/audio and paid media generation. +- Copying reference names, prose, formulas, layouts, source or commercial + assets. +- HTML/gallery implementation in this batch; it begins only after the six + design authorities are integrated, independently reviewed and owner-approved. + +## Human Gates + +- Owner accepts or revises the integrated complete-design direction. +- Owner later reviews every landscape UI page and representative art state. +- Runtime, media production and release remain separately authorized. + +## Open Questions + +None blocking this design batch. Exact simultaneous capacity within the 10-12 +target is selected by the five-profile simulation, not guessed from the +reference demo's observed `x/10` display. diff --git a/.agent-taskgraph/staffing/ABG-V2-DUAL-REVIEW.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-DUAL-REVIEW.md similarity index 100% rename from .agent-taskgraph/staffing/ABG-V2-DUAL-REVIEW.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-DUAL-REVIEW.md diff --git a/.agent-taskgraph/staffing/ABG-V2-EXPERIENCE-REVIEW-R5-A2.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-EXPERIENCE-REVIEW-R5-A2.md similarity index 100% rename from .agent-taskgraph/staffing/ABG-V2-EXPERIENCE-REVIEW-R5-A2.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-EXPERIENCE-REVIEW-R5-A2.md diff --git a/.agent-taskgraph/staffing/ABG-V2-PARALLEL-EXPANSION.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-PARALLEL-EXPANSION.md similarity index 100% rename from .agent-taskgraph/staffing/ABG-V2-PARALLEL-EXPANSION.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-PARALLEL-EXPANSION.md diff --git a/.agent-taskgraph/staffing/ABG-V2-R4-VALIDATOR-H2-A2.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R4-VALIDATOR-H2-A2.md similarity index 100% rename from .agent-taskgraph/staffing/ABG-V2-R4-VALIDATOR-H2-A2.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R4-VALIDATOR-H2-A2.md diff --git a/.agent-taskgraph/staffing/ABG-V2-R4-VALIDATOR-H2-A3.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R4-VALIDATOR-H2-A3.md similarity index 100% rename from .agent-taskgraph/staffing/ABG-V2-R4-VALIDATOR-H2-A3.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R4-VALIDATOR-H2-A3.md diff --git a/.agent-taskgraph/staffing/ABG-V2-R4-VALIDATOR-H2.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R4-VALIDATOR-H2.md similarity index 100% rename from .agent-taskgraph/staffing/ABG-V2-R4-VALIDATOR-H2.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R4-VALIDATOR-H2.md diff --git a/.agent-taskgraph/staffing/ABG-V2-R5-DUAL-REVIEW.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R5-DUAL-REVIEW.md similarity index 100% rename from .agent-taskgraph/staffing/ABG-V2-R5-DUAL-REVIEW.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/ABG-V2-R5-DUAL-REVIEW.md diff --git a/.agent-taskgraph/staffing/experience-r3-repair.md b/.agent-taskgraph/archive/recruit-line-redesign-02/staffing/experience-r3-repair.md similarity index 100% rename from .agent-taskgraph/staffing/experience-r3-repair.md rename to .agent-taskgraph/archive/recruit-line-redesign-02/staffing/experience-r3-repair.md diff --git a/.agent-taskgraph/graph.yaml b/.agent-taskgraph/graph.yaml index 7ff466ce..fbf05a04 100644 --- a/.agent-taskgraph/graph.yaml +++ b/.agent-taskgraph/graph.yaml @@ -1,477 +1,166 @@ version: 1 batch: - id: "recruit-line-redesign-02" - revision: "graph-r11" + id: "abg-p9-1-continuous-development-01" + revision: "graph-p9.1-r1" spec: ".agent-taskgraph/spec.md" - spec_revision: "design-spec-r3" + spec_revision: "p9.1-spec-r1" approved_by: "Owner" - approved_at: "2026-08-10T22:38:28+08:00" - max_concurrency: 6 + approved_at: "2026-08-13" + max_concurrency: 3 nodes: - - id: "v2_core_systems" - title: "Recruit-line core systems redesign" + - id: "p91_game_design" + title: "Design arbitrary-party complete first session" kind: "expert" - goal_ref: "task:ABG-V2-CORE-SYSTEMS" + goal_ref: "task:ABG-P9.1-GAME-DESIGN" needs: [] - consumes: [".agent-taskgraph/spec.md", "docs/reference/"] - produces: ["docs/product/GAME_PRODUCT_CONTRACT.md", "docs/product/SYSTEMS_AND_BATTLE.md", "docs/product/ECONOMY_AND_BALANCE.md", "docs/product/SAVE_AND_FAILURE_CONTRACT.md"] - writes: ["docs/product/"] - acceptance: ["Horizontal ordered-line auto-battle replaces lane grid", "Recruitment/death/dismissal/capacity are executable", "Five pacing profiles and save/failure boundaries are consistent"] - on_pass: "v2_integration" + consumes: [".agent-taskgraph/spec.md", "docs/runtime/PHASE9_0R10_PAIR_ADAPTATION_CONTRACT.md", "docs/product/GAME_PRODUCT_CONTRACT.md"] + produces: ["docs/runtime/p9_1/GAMEPLAY_EXPERIENCE_SPEC.md"] + writes: ["docs/runtime/p9_1/GAMEPLAY_EXPERIENCE_SPEC.md"] + acceptance: ["All six starting pairs complete five battles", "Guild response, route and growth values change with actual party state", "No frozen system or content expansion"] + on_pass: "p91_spec_integration" on_fail: "failed" max_attempts: 2 - - id: "v2_content_matrix" - title: "Generated recruit and complete content matrices" + - id: "p91_ux_visual_design" + title: "Design P9.1 screens, states and causal presentation" kind: "expert" - goal_ref: "task:ABG-V2-CONTENT-MATRIX" + goal_ref: "task:ABG-P9.1-UX-VISUAL" needs: [] - consumes: [".agent-taskgraph/spec.md", "docs/reference/"] - produces: ["docs/content/"] - writes: ["docs/content/"] - acceptance: ["42 professions, 36 traits, 320 equipment, 60 artifacts, 100 enemies, 16 bosses and 8 regions", "No fixed protagonist relationships or filler rows", "All IDs and mechanics validate"] - on_pass: "v2_integration" + consumes: [".agent-taskgraph/spec.md", "docs/runtime/PHASE9_0R10_PAIR_ADAPTATION_CONTRACT.md", "docs/presentation/UI_AND_VISUAL_SYSTEM.md"] + produces: ["docs/runtime/p9_1/UX_VISUAL_FLOW_SPEC.md"] + writes: ["docs/runtime/p9_1/UX_VISUAL_FLOW_SPEC.md"] + acceptance: ["Every P9.1 screen and state has one readable primary action", "Chinese causal copy and landscape hierarchy are explicit", "Motion and accessibility preserve accepted combat feedback"] + on_pass: "p91_spec_integration" on_fail: "failed" max_attempts: 2 - - id: "v2_ux_screens" - title: "Landscape UX and complete screen-state architecture" + - id: "p91_technical_architecture" + title: "Design modular runtime architecture and migration" kind: "expert" - goal_ref: "task:ABG-V2-UX-SCREENS" + goal_ref: "task:ABG-P9.1-TECH-ARCH" needs: [] - consumes: [".agent-taskgraph/spec.md", "docs/reference/"] - produces: ["docs/presentation/SCREEN_AND_STATE_MAP.md", "docs/presentation/UI_AND_VISUAL_SYSTEM.md"] - writes: ["docs/presentation/SCREEN_AND_STATE_MAP.md", "docs/presentation/UI_AND_VISUAL_SYSTEM.md"] - acceptance: ["All player-facing landscape screens and states are covered", "Retired fixed-cast, lane-grid and manual-command UI is absent", "Shop, recruit, equip, reorder, auto-battle and recovery are immediately readable"] - on_pass: "v2_integration" + consumes: [".agent-taskgraph/spec.md", "P9.0R10 runtime architecture baseline", "P9.0R1-R10 test architecture baseline"] + produces: ["docs/runtime/p9_1/RUNTIME_ARCHITECTURE_SPEC.md"] + writes: ["docs/runtime/p9_1/RUNTIME_ARCHITECTURE_SPEC.md"] + acceptance: ["State ownership and APIs are executable", "Extraction preserves P9.0R10 behavior", "Later developer write scopes do not overlap"] + on_pass: "p91_spec_integration" on_fail: "failed" max_attempts: 2 - - id: "v2_art_direction" - title: "Generated-recruit art, animation and VFX production bible" - kind: "expert" - goal_ref: "task:ABG-V2-ART-DIRECTION" - needs: [] - consumes: [".agent-taskgraph/spec.md", "docs/reference/"] - produces: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] - writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] - acceptance: ["Modular recruits and 42 profession identities are production-countable", "Eight regions and 116 enemy identities have readable art/animation coverage", "Horizontal-line combat remains legible at target landscape sizes"] - on_pass: "v2_integration" - on_fail: "failed" - max_attempts: 2 - - - id: "v2_audio_direction" - title: "Music, ambience, SFX and voice production bible" - kind: "expert" - goal_ref: "task:ABG-V2-AUDIO-DIRECTION" - needs: [] - consumes: [".agent-taskgraph/spec.md", "docs/reference/"] - produces: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] - writes: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] - acceptance: ["Audio makes shop, order, battle and failure consequences readable", "Voice scope matches generated recruits and no fixed-cast burden remains", "Production counts, priority, accessibility and lifecycle behavior are explicit"] - on_pass: "v2_integration" - on_fail: "failed" - max_attempts: 2 - - - id: "v2_design_validation" - title: "Pre-development design validator and coverage contract" - kind: "expert" - goal_ref: "task:ABG-V2-DESIGN-VALIDATION" - needs: [] - consumes: [".agent-taskgraph/spec.md", "docs/reference/"] - produces: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] - writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] - acceptance: ["Expected counts and stable IDs are mechanically checkable", "Retired concepts and cross-reference failures are detected", "Validator reports unknown pending bindings without inventing product authority"] - on_pass: "v2_integration" - on_fail: "failed" - max_attempts: 2 - - - id: "v2_integration" - title: "Integrate and reconcile the redesign" + - id: "p91_spec_integration" + title: "Integrate and freeze P9.1 implementation contract" kind: "merge" - goal_ref: "task:ABG-V2-INTEGRATION" - needs: ["v2_core_systems", "v2_content_matrix", "v2_ux_screens", "v2_art_direction", "v2_audio_direction", "v2_design_validation"] - consumes: ["docs/product/", "docs/content/", "docs/presentation/", "docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] - produces: ["clean main revision", "contract validation report"] - writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] - acceptance: ["No cross-document contradiction", "Repository validator passes", "All six design authorities are integrated", "Fixed revision ready for independent review"] - on_pass: "v2_systems_review" - on_fail: "failed" - max_attempts: 1 - - id: "v2_systems_review" - title: "Independent systems integrity review" - kind: "verifier" - goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R2" - needs: ["v2_integration"] - consumes: ["fixed integrated revision"] - produces: ["read-only systems PASS/FAIL/INVALID review"] - writes: [] - acceptance: ["Core loop, battle, economy, pacing, save/failure and mechanical content coherence pass"] - on_pass: "v2_review_synthesis" - on_fail: "v2_battle_outcome_repair" - max_attempts: 1 - - - id: "v2_experience_review" - title: "Independent experience and production review" - kind: "verifier" - goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R2" - needs: ["v2_integration"] - consumes: ["fixed integrated revision"] - produces: ["read-only experience PASS/FAIL/INVALID review"] - writes: [] - acceptance: ["UX comprehension, content presentation, art/animation/audio feasibility, accessibility and originality pass"] - on_pass: "v2_review_synthesis" + goal_ref: "task:ABG-P9.1-SPEC-INTEGRATION" + needs: ["p91_game_design", "p91_ux_visual_design", "p91_technical_architecture"] + consumes: ["docs/runtime/p9_1/GAMEPLAY_EXPERIENCE_SPEC.md", "docs/runtime/p9_1/UX_VISUAL_FLOW_SPEC.md", "docs/runtime/p9_1/RUNTIME_ARCHITECTURE_SPEC.md"] + produces: ["docs/runtime/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_CONTRACT.md"] + writes: ["docs/runtime/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_CONTRACT.md"] + acceptance: ["Product, presentation and architecture terms agree", "All implementation values and boundaries are frozen", "No unresolved decision is delegated to a developer"] + on_pass: "p91_runtime_foundation" on_fail: "failed" max_attempts: 1 - - id: "v2_battle_outcome_repair" - title: "Define deterministic battle outcome vocabulary" - kind: "expert" - goal_ref: "task:ABG-V2-BATTLE-OUTCOME-REPAIR" - needs: ["v2_systems_review"] - consumes: ["systems R2 FAIL", "docs/product/SYSTEMS_AND_BATTLE.md", "all current miss/dodge/evasion references"] - produces: ["one executable deterministic outcome contract"] - writes: ["docs/product/SYSTEMS_AND_BATTLE.md"] - acceptance: ["No accuracy/evasion/hidden-miss probability is introduced", "Every applied/invalidated/resisted/immune/intercepted outcome is executable and journaled", "Skill schema can bind deterministic outcome triggers"] - on_pass: "v2_content_outcome_sync" + - id: "p91_runtime_foundation" + title: "Extract behavior-equivalent runtime foundations" + kind: "worker" + goal_ref: "task:ABG-P9.1-RUNTIME-FOUNDATION" + needs: ["p91_spec_integration"] + consumes: ["docs/runtime/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_CONTRACT.md", "P9.0R10 fixed runtime source"] + produces: ["runtime/domain/foundation/", "runtime/ui/foundation/"] + writes: ["runtime/main.gd", "runtime/domain/foundation/", "runtime/ui/foundation/"] + acceptance: ["P9.0R1-R10 declared behavior remains equivalent", "Shared state has one owner", "Parallel extension APIs compile and are covered"] + on_pass: "p91_domain_implementation" on_fail: "failed" - max_attempts: 1 + max_attempts: 2 - - id: "v2_content_outcome_sync" - title: "Synchronize profession and content outcome semantics" - kind: "expert" - goal_ref: "task:ABG-V2-CONTENT-OUTCOME-SYNC" - needs: ["v2_battle_outcome_repair"] - consumes: ["deterministic outcome contract"] - produces: ["content rows bound to declared outcomes"] - writes: ["docs/content/PROFESSION_CATALOG.md", "docs/content/EQUIPMENT_CATALOG.md", "docs/content/TRAIT_CATALOG.md", "docs/content/ENEMY_CATALOG.md"] - acceptance: ["PF-A04 is executable without accuracy/evasion", "All miss/dodge wording maps to a declared outcome or is rewritten", "No content count or unrelated row changes"] - on_pass: "v2_outcome_reintegration" + - id: "p91_domain_implementation" + title: "Implement arbitrary-party P9.1 domain flow" + kind: "worker" + goal_ref: "task:ABG-P9.1-DOMAIN" + needs: ["p91_runtime_foundation"] + consumes: ["runtime/domain/foundation/", "docs/runtime/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_CONTRACT.md"] + produces: ["runtime/domain/p9_1/"] + writes: ["runtime/domain/p9_1/"] + acceptance: ["Six-pair deterministic state matrix passes", "Transactions commit exactly once", "No Control-tree or frame-rate dependency"] + on_pass: "p91_runtime_integration" on_fail: "failed" - max_attempts: 1 + max_attempts: 2 - - id: "v2_art_outcome_sync" - title: "Synchronize art and VFX outcome semantics" - kind: "expert" - goal_ref: "task:ABG-V2-ART-OUTCOME-SYNC" - needs: ["v2_battle_outcome_repair"] - consumes: ["deterministic outcome contract"] - produces: ["art/VFX cues bound to declared outcomes"] - writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] - acceptance: ["Avoidance cues name only declared deterministic outcomes", "Readability and reduced-motion variants remain explicit"] - on_pass: "v2_outcome_reintegration" + - id: "p91_ui_implementation" + title: "Implement P9.1 Godot screens and feedback" + kind: "worker" + goal_ref: "task:ABG-P9.1-UI" + needs: ["p91_runtime_foundation"] + consumes: ["runtime/ui/foundation/", "docs/runtime/p9_1/UX_VISUAL_FLOW_SPEC.md"] + produces: ["runtime/ui/p9_1/"] + writes: ["runtime/ui/p9_1/"] + acceptance: ["All declared screens render from real state", "Chinese first and accessibility variants fit", "Accepted battle movement and feedback are preserved"] + on_pass: "p91_runtime_integration" on_fail: "failed" - max_attempts: 1 + max_attempts: 2 - - id: "v2_audio_outcome_sync" - title: "Synchronize audio outcome semantics" - kind: "expert" - goal_ref: "task:ABG-V2-AUDIO-OUTCOME-SYNC" - needs: ["v2_battle_outcome_repair"] - consumes: ["deterministic outcome contract"] - produces: ["SFX/caption cues bound to declared outcomes"] - writes: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] - acceptance: ["Dodge/miss cues name only declared deterministic outcomes", "Bindings, production counts and accessibility remain valid"] - on_pass: "v2_outcome_reintegration" + - id: "p91_test_implementation" + title: "Implement P9.1 tests, matrix and capture harness" + kind: "worker" + goal_ref: "task:ABG-P9.1-TESTS" + needs: ["p91_runtime_foundation"] + consumes: ["docs/runtime/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_CONTRACT.md", "runtime/tests/"] + produces: ["runtime/tests/p9_1/", "runtime/tools/p9_1/"] + writes: ["runtime/tests/p9_1/", "runtime/tools/p9_1/"] + acceptance: ["Pair-path and negative boundaries are executable", "Parsed touch and layout states are declared", "Strict diagnostics and cleanup are enforced"] + on_pass: "p91_runtime_integration" on_fail: "failed" - max_attempts: 1 + max_attempts: 2 - - id: "v2_outcome_reintegration" - title: "Reintegrate deterministic battle outcome repair" + - id: "p91_runtime_integration" + title: "Integrate the complete P9.1 first session" kind: "merge" - goal_ref: "task:ABG-V2-OUTCOME-REINTEGRATION" - needs: ["v2_battle_outcome_repair", "v2_content_outcome_sync", "v2_art_outcome_sync", "v2_audio_outcome_sync"] - consumes: ["repaired product, content, art and audio authorities"] - produces: ["new fixed review revision", "validator and semantic reference audit"] - writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] - acceptance: ["Validator remains green", "No undeclared accuracy/evasion/miss mechanic remains", "R2 FAIL and INVALID are append-only history"] - on_pass: "v2_systems_review_r3" + goal_ref: "task:ABG-P9.1-RUNTIME-INTEGRATION" + needs: ["p91_domain_implementation", "p91_ui_implementation", "p91_test_implementation"] + consumes: ["runtime/domain/p9_1/", "runtime/ui/p9_1/", "runtime/tests/p9_1/", "runtime/tools/p9_1/"] + produces: ["P9.1 fixed candidate revision"] + writes: ["runtime/main.gd", "runtime/project.godot", "runtime/tools/run_p9_1_tests.sh"] + acceptance: ["Complete five-battle flow works for all pairs", "Full R1-R10 plus P9.1 suite passes", "Normal-render and touch evidence meet the frozen contract"] + on_pass: "p91_independent_review" on_fail: "failed" - max_attempts: 1 + max_attempts: 2 - - id: "v2_systems_review_r3" - title: "Fresh systems integrity re-review" + - id: "p91_independent_review" + title: "Independently review the fixed P9.1 candidate" kind: "verifier" - goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R3" - needs: ["v2_outcome_reintegration"] - consumes: ["new fixed integrated revision"] - produces: ["read-only systems PASS/FAIL/INVALID terminal"] + goal_ref: "task:ABG-P9.1-INDEPENDENT-REVIEW" + needs: ["p91_runtime_integration"] + consumes: ["P9.1 fixed candidate revision", "docs/runtime/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_CONTRACT.md"] + produces: ["read-only PASS, FAIL or INVALID terminal"] writes: [] - acceptance: ["All systems gates rerun from zero, including deterministic outcome vocabulary"] - on_pass: "v2_review_synthesis" - on_fail: "v2_anti_dominance_simulation_repair" - max_attempts: 1 - - - id: "v2_anti_dominance_simulation_repair" - title: "Add executable fixed-seed anti-dominance fixtures" - kind: "expert" - goal_ref: "task:ABG-V2-ANTI-DOMINANCE-SIMULATION-REPAIR" - needs: ["v2_systems_review_r3"] - consumes: ["Systems R3 Gate 6 FAIL", "economy/build policy authority"] - produces: ["five exact seed fixtures", "executable policy comparison", "documented counterfactual outcomes"] - writes: ["docs/product/ECONOMY_AND_BALANCE.md", "docs/product/fixtures/ANTI_DOMINANCE_FIXED_SEEDS.json", "tools/simulate_anti_dominance.py"] - acceptance: ["Five fixed seeds and all inputs are reproducible", "No policy dominates all legal states", "Self-test rejects a deliberately dominant fixture"] - on_pass: "v2_experience_reintegration" + acceptance: ["Review starts from a clean immutable revision", "Every gate is rerun from zero", "Reviewer makes no repair or product write"] + on_pass: "p91_h5_publish" on_fail: "failed" max_attempts: 1 - - id: "v2_systems_review_r4" - title: "Fresh systems integrity re-review after anti-dominance repair" - kind: "verifier" - goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R4" - needs: ["v2_experience_reintegration"] - consumes: ["new immutable repaired revision"] - produces: ["read-only systems PASS/FAIL/INVALID terminal"] - writes: [] - acceptance: ["All systems gates rerun from zero including fixed-seed anti-dominance fixtures"] - on_pass: "v2_review_synthesis" + - id: "p91_h5_publish" + title: "Export and publish the accepted P9.1 H5" + kind: "worker" + goal_ref: "task:ABG-P9.1-H5-PUBLISH" + needs: ["p91_independent_review"] + consumes: ["P9.1 fixed candidate revision", "read-only PASS, FAIL or INVALID terminal"] + produces: ["new private Artifact Gallery item"] + writes: ["runtime/build/web-preview/"] + acceptance: ["Runtime-only export excludes tests and evidence", "Remote PCK is source-bound", "Remote Chinese Title loads with zero browser diagnostics"] + on_pass: "p91_owner_play_gate" on_fail: "failed" - max_attempts: 1 + max_attempts: 2 - - id: "v2_experience_review_r3" - title: "Fresh experience and production re-review" - kind: "verifier" - goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R3" - needs: ["v2_outcome_reintegration"] - consumes: ["new fixed integrated revision", "actual reference authority paths"] - produces: ["read-only experience PASS/FAIL/INVALID terminal"] - writes: [] - acceptance: ["All experience gates rerun from zero with valid context paths"] - on_pass: "v2_review_synthesis" - on_fail: "v2_experience_repair_fanout" - max_attempts: 1 - - - id: "v2_experience_repair_fanout" - title: "Repair R3 experience cross-document failures" - kind: "fork" - goal_ref: "decision:ABG-V2-EXPERIENCE-R3-REPAIR" - needs: ["v2_experience_review_r3"] - consumes: ["R3 experience FAIL terminal"] - produces: ["five disjoint repair candidates"] - writes: [] - acceptance: ["Every reported contradiction has one owner and non-overlapping write scope"] - on_pass: "v2_experience_reintegration" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_onboarding_presentation_repair" - title: "Repair run creation, onboarding and party-line presentation" - kind: "expert" - goal_ref: "task:ABG-V2-ONBOARDING-PRESENTATION-REPAIR" - needs: ["v2_experience_review_r3"] - consumes: ["product run-creation authority", "R3 experience FAIL"] - produces: ["complete provisional-recruit decision state and one party-line order"] - writes: ["docs/presentation/SCREEN_AND_STATE_MAP.md", "docs/presentation/UI_AND_VISUAL_SYSTEM.md"] - acceptance: ["Four provisional offers and choose-two revision are explicit before run creation", "First exposed position has one cross-document direction", "Screen count remains intentional and validator-readable"] - on_pass: "v2_experience_reintegration" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_art_schema_order_repair" - title: "Repair recruit visual schema mapping and party-line order" - kind: "expert" - goal_ref: "task:ABG-V2-ART-SCHEMA-ORDER-REPAIR" - needs: ["v2_experience_review_r3"] - consumes: ["recruit generation schema", "UI party-line authority", "R3 experience FAIL"] - produces: ["exact content-to-art recipe crosswalk and coherent counts"] - writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] - acceptance: ["Every recruit content field maps to an art field or explicit derived assembly", "Counts and totals recalculate", "Party line matches UI authority"] - on_pass: "v2_experience_reintegration" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_audio_coverage_repair" - title: "Repair screen and abandonment audio coverage" - kind: "expert" - goal_ref: "task:ABG-V2-AUDIO-COVERAGE-REPAIR" - needs: ["v2_experience_review_r3"] - consumes: ["50-screen authority", "OUT-006 abandonment state", "R3 experience FAIL"] - produces: ["complete screen/action audio binding"] - writes: ["docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md"] - acceptance: ["OUT-006 is explicitly bound", "Screen total is 50", "Audio production totals remain internally consistent"] - on_pass: "v2_experience_reintegration" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_reference_scope_repair" - title: "Repair transferred-lesson scope to generated recruits" - kind: "expert" - goal_ref: "task:ABG-V2-REFERENCE-SCOPE-REPAIR" - needs: ["v2_experience_review_r3"] - consumes: ["design-spec-r3", "current product/content authorities", "R3 experience FAIL"] - produces: ["current original-scope lessons without fixed-cast residue"] - writes: ["docs/reference/TRANSFERABLE_STRUCTURE_LESSONS.md"] - acceptance: ["No named-recruit or relationship requirement remains", "Current generated-recruit and profession scope is exact", "Reference facts remain separate from original design commitments"] - on_pass: "v2_experience_reintegration" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_crossdoc_validator_repair" - title: "Add regression gates for R3 experience contradictions" - kind: "expert" - goal_ref: "task:ABG-V2-CROSSDOC-VALIDATOR-REPAIR" - needs: ["v2_experience_review_r3"] - consumes: ["five R3 contradiction classes", "current design validator"] - produces: ["self-tested cross-document regression gates"] - writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] - acceptance: ["Each R3 contradiction class has a positive and negative check", "Existing exact counts remain enforced", "Self-test and full validation pass only after all repair candidates integrate"] - on_pass: "v2_experience_reintegration" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_experience_reintegration" - title: "Integrate and freeze R3 experience repairs" - kind: "merge" - goal_ref: "task:ABG-V2-EXPERIENCE-REINTEGRATION" - needs: ["v2_onboarding_presentation_repair", "v2_art_schema_order_repair", "v2_audio_coverage_repair", "v2_reference_scope_repair", "v2_crossdoc_validator_repair", "v2_anti_dominance_simulation_repair"] - consumes: ["five repair candidates", "systems R3 terminal"] - produces: ["new immutable review revision"] - writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] - acceptance: ["All R3 experience blockers are closed", "Systems terminal is reconciled", "Validator self-test/full/JSON pass"] - on_pass: "v2_experience_review_r4" - on_fail: "v2_crossdoc_validator_h1" - max_attempts: 1 - - - id: "v2_crossdoc_validator_h1" - title: "Repair validator integration against real repaired authorities" - kind: "expert" - goal_ref: "task:ABG-V2-CROSSDOC-VALIDATOR-H1" - needs: ["v2_experience_reintegration"] - consumes: ["integrated main false-negative diagnostics", "real repaired UX and art authorities"] - produces: ["real-document parser repair", "positive and negative regression fixtures"] - writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] - acceptance: ["Integrated repaired main passes text and JSON", "20-case self-test remains strict", "Each R3 contradiction mutation still fails independently"] - on_pass: "v2_experience_reintegration" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_experience_review_r4" - title: "Fresh experience and production re-review after R3 repair" - kind: "verifier" - goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R4" - needs: ["v2_experience_reintegration"] - consumes: ["new immutable repaired revision"] - produces: ["read-only experience PASS/FAIL/INVALID terminal"] - writes: [] - acceptance: ["All experience gates rerun from zero; no R3 evidence reused as PASS"] - on_pass: "v2_review_synthesis" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_review_synthesis" - title: "PMO dual-review synthesis" - kind: "merge" - goal_ref: "task:ABG-V2-REVIEW-SYNTHESIS" - needs: ["v2_systems_review_r4", "v2_experience_review_r4"] - consumes: ["systems review terminal", "experience review terminal"] - produces: ["one PMO acceptance or earliest repair route"] - writes: [".agent-taskgraph/queue/review/", ".agent-taskgraph/STATUS.md"] - acceptance: ["Both reviewers used the same fixed revision", "No open reviewer failure or invalid environment", "Open human and implementation gates remain explicit"] - on_pass: "owner_v2_gate" - on_fail: "failed" - max_attempts: 1 - - - id: "owner_v2_gate" - title: "Owner approves complete design before HTML prototype redesign" + - id: "p91_owner_play_gate" + title: "Owner judges the complete first session" kind: "human_gate" - goal_ref: "decision:ABG-V2-OWNER-GATE" - needs: ["v2_review_synthesis_r5"] - consumes: ["integrated design", "independent review"] - produces: ["owner accept or revise decision"] + goal_ref: "decision:ABG-P9.1-OWNER-PLAY" + needs: ["p91_h5_publish"] + consumes: ["new private Artifact Gallery item"] + produces: ["Owner accept or revise decision"] writes: [] - acceptance: ["Owner authorizes landscape HTML prototype redesign"] + acceptance: ["Owner understands the choices and consequences", "Owner judges the first session worth continuing"] on_pass: "done" on_fail: "failed" max_attempts: 1 - - - id: "v2_r4_seed_transaction_repair" - title: "Repair provisional Recruit seed and transaction authority" - kind: "expert" - goal_ref: "task:ABG-V2-RUN-CREATION-TRANSACTION-REPAIR" - needs: ["v2_systems_review_r4"] - consumes: ["Systems R4 Gate 3 FAIL", "run creation, save and recruit generation authorities"] - produces: ["durable pre-run seed allocation", "reproducible offer receipt", "atomic current_run commit and retry rules"] - writes: ["docs/product/GAME_PRODUCT_CONTRACT.md", "docs/product/SAVE_AND_FAILURE_CONTRACT.md", "docs/content/RECRUIT_GENERATION.md"] - acceptance: ["The seed exists durably before offers", "Crash and retry reproduce identical offers/selection", "Run creation commits once without free reroll or duplicate mutation"] - on_pass: "v2_r4_reintegration" - on_fail: "failed" - max_attempts: 2 - - - id: "v2_r4_hidden_profession_art_repair" - title: "Repair hidden profession content-to-art bindings" - kind: "expert" - goal_ref: "task:ABG-V2-HIDDEN-PROFESSION-ART-REPAIR" - needs: ["v2_experience_review_r4"] - consumes: ["Experience R4 Gate 7 FAIL", "PF-H01..H06 content production briefs"] - produces: ["exact hidden-profession parent, silhouette, prop and motion bindings"] - writes: ["docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md"] - acceptance: ["PF-H01..H06 match content mechanics and production briefs", "No unresolved parent binding remains", "Art totals and other profession rows stay unchanged"] - on_pass: "v2_r4_reintegration" - on_fail: "failed" - max_attempts: 2 - - - id: "v2_r4_reintegration" - title: "Integrate R4 bounded repairs" - kind: "merge" - goal_ref: "task:ABG-V2-R4-REINTEGRATION" - needs: ["v2_r4_seed_transaction_repair", "v2_r4_hidden_profession_art_repair"] - consumes: ["two disjoint repair candidates"] - produces: ["one integrated R4-repaired revision"] - writes: ["docs/reviews/PRE_DEVELOPMENT_READINESS_AUDIT.md"] - acceptance: ["Both exact failure boundaries close", "Existing validators and anti-dominance replay remain green"] - on_pass: "v2_r4_validator_h2" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_r4_validator_h2" - title: "Add R4 contradiction regression gates" - kind: "expert" - goal_ref: "task:ABG-V2-R4-VALIDATOR-H2" - needs: ["v2_r4_reintegration"] - consumes: ["integrated repaired authorities", "both R4 FAIL reports"] - produces: ["positive parser coverage and isolated negative mutations for both contradictions"] - writes: ["docs/reviews/DESIGN_VALIDATION_CONTRACT.md", "tools/validate_predevelopment_design.py"] - acceptance: ["Real repaired documents pass", "Seed-order/transaction mutation fails", "Hidden-profession art mismatch/unresolved mutation fails"] - on_pass: "v2_systems_review_r5" - on_fail: "failed" - max_attempts: 2 - - - id: "v2_systems_review_r5" - title: "Fresh systems integrity review after R4 repairs" - kind: "verifier" - goal_ref: "task:ABG-V2-SYSTEMS-REVIEW-R5" - needs: ["v2_r4_validator_h2"] - consumes: ["new immutable R5 review source"] - produces: ["read-only systems PASS/FAIL/INVALID terminal"] - writes: [] - acceptance: ["All systems gates rerun from zero including provisional seed transaction"] - on_pass: "v2_experience_review_r5" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_experience_review_r5" - title: "Fresh experience and production review after R4 repairs" - kind: "verifier" - goal_ref: "task:ABG-V2-EXPERIENCE-REVIEW-R5" - needs: ["v2_r4_validator_h2"] - consumes: ["new immutable R5 review source"] - produces: ["read-only experience PASS/FAIL/INVALID terminal"] - writes: [] - acceptance: ["All experience gates rerun from zero including hidden-profession production binding"] - on_pass: "v2_review_synthesis_r5" - on_fail: "failed" - max_attempts: 1 - - - id: "v2_review_synthesis_r5" - title: "PMO R5 dual-review synthesis" - kind: "merge" - goal_ref: "task:ABG-V2-REVIEW-SYNTHESIS-R5" - needs: ["v2_systems_review_r5", "v2_experience_review_r5"] - consumes: ["systems R5 terminal", "experience R5 terminal"] - produces: ["one PMO acceptance or earliest repair route"] - writes: [".agent-taskgraph/queue/review/", ".agent-taskgraph/STATUS.md"] - acceptance: ["Both reviewers used the same fixed revision", "Both terminals PASS", "Human and implementation gates remain explicit"] - on_pass: "owner_v2_gate" - on_fail: "failed" - max_attempts: 1 diff --git a/.agent-taskgraph/queue/active/.gitkeep b/.agent-taskgraph/queue/active/.gitkeep index e69de29b..8b137891 100644 --- a/.agent-taskgraph/queue/active/.gitkeep +++ b/.agent-taskgraph/queue/active/.gitkeep @@ -0,0 +1 @@ + diff --git a/.agent-taskgraph/queue/done/.gitkeep b/.agent-taskgraph/queue/done/.gitkeep index e69de29b..8b137891 100644 --- a/.agent-taskgraph/queue/done/.gitkeep +++ b/.agent-taskgraph/queue/done/.gitkeep @@ -0,0 +1 @@ + diff --git a/.agent-taskgraph/queue/failed/.gitkeep b/.agent-taskgraph/queue/failed/.gitkeep index e69de29b..8b137891 100644 --- a/.agent-taskgraph/queue/failed/.gitkeep +++ b/.agent-taskgraph/queue/failed/.gitkeep @@ -0,0 +1 @@ + diff --git a/.agent-taskgraph/queue/inbox/.gitkeep b/.agent-taskgraph/queue/inbox/.gitkeep index e69de29b..8b137891 100644 --- a/.agent-taskgraph/queue/inbox/.gitkeep +++ b/.agent-taskgraph/queue/inbox/.gitkeep @@ -0,0 +1 @@ + diff --git a/.agent-taskgraph/queue/review/.gitkeep b/.agent-taskgraph/queue/review/.gitkeep index e69de29b..8b137891 100644 --- a/.agent-taskgraph/queue/review/.gitkeep +++ b/.agent-taskgraph/queue/review/.gitkeep @@ -0,0 +1 @@ + diff --git a/.agent-taskgraph/roles/p91-domain-developer/ROLE.md b/.agent-taskgraph/roles/p91-domain-developer/ROLE.md new file mode 100644 index 00000000..1512eb06 --- /dev/null +++ b/.agent-taskgraph/roles/p91-domain-developer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 领域逻辑开发 + +| 字段 | 值 | +|---|---| +| Role ID | p91-domain-developer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | available | +| 核心职责 | Implement deterministic arbitrary-party session state and transactions | +| 负责范围 | `runtime/domain/p9_1/` after foundation integration | +| 明确不负责 | Shared entry, UI, PMO state, packaging or release | +| 当前 Goal | PENDING | +| 当前 Session ID | PENDING | +| 连续性记录 | Reserved for wave three | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-game-designer/ROLE.md b/.agent-taskgraph/roles/p91-game-designer/ROLE.md new file mode 100644 index 00000000..b3494655 --- /dev/null +++ b/.agent-taskgraph/roles/p91-game-designer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 游戏策划 + +| 字段 | 值 | +|---|---| +| Role ID | p91-game-designer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | reserved | +| 核心职责 | Define arbitrary-party first-session decisions, pacing, consequences and state matrix | +| 负责范围 | `docs/runtime/p9_1/GAMEPLAY_EXPERIENCE_SPEC.md` | +| 明确不负责 | Runtime code, UI implementation, final assets, packaging or PMO state | +| 当前 Goal | task:ABG-P9.1-GAME-DESIGN | +| 当前 Session ID | PENDING | +| 连续性记录 | New Phase 9 runtime design role | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-h5-publisher/ROLE.md b/.agent-taskgraph/roles/p91-h5-publisher/ROLE.md new file mode 100644 index 00000000..576945c1 --- /dev/null +++ b/.agent-taskgraph/roles/p91-h5-publisher/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 H5 发布员 + +| 字段 | 值 | +|---|---| +| Role ID | p91-h5-publisher | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | task-scoped | +| 状态 | available | +| 核心职责 | Export reviewer-approved runtime-only Web build and publish a private Gallery item | +| 负责范围 | Ignored source-bound Web output and Gallery delivery | +| 明确不负责 | Product repair, mobile packaging, TestFlight, devices, stores or release | +| 当前 Goal | PENDING | +| 当前 Session ID | PENDING | +| 连续性记录 | Fresh session required after reviewer PASS | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-independent-reviewer/ROLE.md b/.agent-taskgraph/roles/p91-independent-reviewer/ROLE.md new file mode 100644 index 00000000..48cc468b --- /dev/null +++ b/.agent-taskgraph/roles/p91-independent-reviewer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 独立 Reviewer + +| 字段 | 值 | +|---|---| +| Role ID | p91-independent-reviewer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | task-scoped | +| 状态 | available | +| 核心职责 | Review one immutable candidate from zero and report PASS, FAIL or INVALID | +| 负责范围 | Read-only candidate and reviewer-owned ignored evidence | +| 明确不负责 | Repair, commit, push, product writes, packaging or release | +| 当前 Goal | PENDING | +| 当前 Session ID | PENDING | +| 连续性记录 | Fresh session required after integration | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-integration-developer/ROLE.md b/.agent-taskgraph/roles/p91-integration-developer/ROLE.md new file mode 100644 index 00000000..57c8ae73 --- /dev/null +++ b/.agent-taskgraph/roles/p91-integration-developer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 运行时集成开发 + +| 字段 | 值 | +|---|---| +| Role ID | p91-integration-developer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | available | +| 核心职责 | Integrate parallel runtime outputs and own shared entry/regression closure | +| 负责范围 | Goal-defined shared runtime entry and runner files | +| 明确不负责 | Unapproved redesign, PMO state, packaging or release | +| 当前 Goal | PENDING | +| 当前 Session ID | PENDING | +| 连续性记录 | Reserved for integration wave | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-runtime-architect/ROLE.md b/.agent-taskgraph/roles/p91-runtime-architect/ROLE.md new file mode 100644 index 00000000..603bb66f --- /dev/null +++ b/.agent-taskgraph/roles/p91-runtime-architect/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 技术架构师 + +| 字段 | 值 | +|---|---| +| Role ID | p91-runtime-architect | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | reserved | +| 核心职责 | Define state ownership, module APIs, extraction order, parallel write boundaries and regression seams | +| 负责范围 | `docs/runtime/p9_1/RUNTIME_ARCHITECTURE_SPEC.md` | +| 明确不负责 | Product behavior changes, runtime implementation, assets, packaging or PMO state | +| 当前 Goal | task:ABG-P9.1-TECH-ARCH | +| 当前 Session ID | PENDING | +| 连续性记录 | New Phase 9 runtime architecture role | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-runtime-foundation-developer/ROLE.md b/.agent-taskgraph/roles/p91-runtime-foundation-developer/ROLE.md new file mode 100644 index 00000000..d1ea5391 --- /dev/null +++ b/.agent-taskgraph/roles/p91-runtime-foundation-developer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 运行时基础开发 + +| 字段 | 值 | +|---|---| +| Role ID | p91-runtime-foundation-developer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | available | +| 核心职责 | Extract behavior-equivalent shared runtime state and extension APIs | +| 负责范围 | Goal-defined foundation paths and shared entry during its exclusive stage | +| 明确不负责 | Product redesign, PMO state, packaging or release | +| 当前 Goal | PENDING | +| 当前 Session ID | PENDING | +| 连续性记录 | Reserved for wave two | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-test-developer/ROLE.md b/.agent-taskgraph/roles/p91-test-developer/ROLE.md new file mode 100644 index 00000000..8a369174 --- /dev/null +++ b/.agent-taskgraph/roles/p91-test-developer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 测试与证据开发 + +| 字段 | 值 | +|---|---| +| Role ID | p91-test-developer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | available | +| 核心职责 | Implement P9.1 matrices, mobile input, layout and capture harnesses | +| 负责范围 | `runtime/tests/p9_1/`, `runtime/tools/p9_1/` | +| 明确不负责 | Product code, PMO state, packaging or release | +| 当前 Goal | PENDING | +| 当前 Session ID | PENDING | +| 连续性记录 | Reserved for wave three | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-ui-developer/ROLE.md b/.agent-taskgraph/roles/p91-ui-developer/ROLE.md new file mode 100644 index 00000000..cc44758e --- /dev/null +++ b/.agent-taskgraph/roles/p91-ui-developer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 Godot UI 开发 + +| 字段 | 值 | +|---|---| +| Role ID | p91-ui-developer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | available | +| 核心职责 | Implement real Godot P9.1 screens and feedback from declared state snapshots | +| 负责范围 | `runtime/ui/p9_1/` after foundation integration | +| 明确不负责 | Domain formulas, shared entry, PMO state, packaging or release | +| 当前 Goal | PENDING | +| 当前 Session ID | PENDING | +| 连续性记录 | Reserved for wave three | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/roles/p91-ux-visual-designer/ROLE.md b/.agent-taskgraph/roles/p91-ux-visual-designer/ROLE.md new file mode 100644 index 00000000..79634e7f --- /dev/null +++ b/.agent-taskgraph/roles/p91-ux-visual-designer/ROLE.md @@ -0,0 +1,16 @@ +# Role: P9.1 UX/视觉设计师 + +| 字段 | 值 | +|---|---| +| Role ID | p91-ux-visual-designer | +| Team revision | rev-p9.1-1 | +| Origin | p9.1-spec-r1 | +| 生命周期 | persistent | +| 状态 | reserved | +| 核心职责 | Define exact landscape screens, Chinese causal copy, hierarchy, motion and accessibility | +| 负责范围 | `docs/runtime/p9_1/UX_VISUAL_FLOW_SPEC.md` | +| 明确不负责 | Product formulas, runtime code, paid/final media, packaging or PMO state | +| 当前 Goal | task:ABG-P9.1-UX-VISUAL | +| 当前 Session ID | PENDING | +| 连续性记录 | New Phase 9 runtime presentation role | +| 最后更新 | 2026-08-13 / PMO | diff --git a/.agent-taskgraph/spec.md b/.agent-taskgraph/spec.md index db591b22..7ce4eb87 100644 --- a/.agent-taskgraph/spec.md +++ b/.agent-taskgraph/spec.md @@ -1,150 +1,80 @@ -# Execution Spec: Recruit-Line Auto-Battler Redesign +# Execution Spec: P9.1 Arbitrary-Party Complete First Session > Status: FROZEN -> Requested: 2026-08-10 -> Frozen by: Owner / 2026-08-10 -> Revision: design-spec-r3 +> Requested: 2026-08-13 +> Frozen by: Owner / 2026-08-13 +> Revision: p9.1-spec-r1 -## Decision This Batch Enables +## 本批要回答的问题 -Produce a coherent replacement for the current fixed-cast, six-unit, -three-lane design and, in parallel, replace the obsolete presentation -authorities with implementation-ready UX, art, audio and validation contracts. -This batch still does not start runtime development or final media production. +P9.0R10 允许玩家从四人中任意选择两人，并让全部六种无序组合完成前两战。 +但旧 R5-R7 只完整支持常规“前排 + 远程/治疗”结构。P9.1 必须让每一种合法 +双人组合都完成同一条可理解、有取舍、有因果结果的首局。 -## Confirmed Product Direction +## 冻结玩家链路 -- Original landscape 2D Japanese hand-drawn fantasy auto-battler. -- The transferable loop is: shop offers -> recruit/buy -> equip -> reorder one - horizontal party line -> choose risk -> fully automatic battle -> reward, - death, dismissal or replacement -> repeat -> persistent run progression. -- Recruits are compositional units built from generated name, modular visual - identity, fixed profession, rolled attributes, one lightweight mechanical - trait, equipment, level and survival history. -- Recruits are not 30 fixed protagonists. Remove relationship, affinity, - romance, personal-arc and mandatory individual voice systems. -- Battle uses one ordered horizontal line. Remove the 3-lane x front/rear grid, - in-battle Directive system, Anchor/Cache battle terminology and fixed six-unit - cap from the core loop. -- Battle is automated. Player controls are prepare, reorder, equip, start, - pause, speed, inspect and retreat; no repeated manual skill activation. -- Simultaneous deployment starts small and grows toward a simulated target of - 10-12. Cumulative recruits in a run are not capped by that deployment number; - death, dismissal and replacement keep recruitment relevant. +每一种双人组合都必须可达： -## Complete Content Envelope +`任选两人 -> 系统排阵 -> 第一战 -> 组合诊断 -> 安全/节奏适配 -> 第二战 ->` +`公会响应 -> 第三战 -> 路线选择 -> 第四战 -> 个人成长 -> 第五战 -> 首局结果` -- 12 base professions, 24 regular advanced professions and 6 hidden hybrid - professions: 42 playable profession identities. -- 36 lightweight mechanical recruit traits. -- 320 equipment items and 60 artifacts. -- 100 ordinary/elite enemy identities plus 16 bosses across 8 regions. -- First clear target 15-25 hours; mastery and high-difficulty play 60+ hours. -- Minimal world framing and regional progression only. Content depth comes from - profession, equipment, formation-order, enemy and run-economy interaction. +公会响应、路线与成长必须读取实际队伍状态；不得把异常组合套入旧的常规组合结果。 -## Workstreams +## 体验约束 -### ABG-V2-CORE-SYSTEMS +- 玩家只选人，不手动选前后排；系统按角色能力形成横向队列并解释结果。 +- 第一个有意义选择不能预选；至少两个选项在不同组合或状态下各自合理。 +- 第三人、保留装备、路线覆盖和个人成长之间必须有真实机会成本或能力变化。 +- 每战保留近战碰撞、远程攻击、治疗/保护、伤害数字和结果因果的可读性。 +- 结果页回答：发生了什么、为什么、哪个决定造成差异、下一步是什么。 +- 中文是第一审查语言；英文、130% 文本、降低动态、静音和低功耗不得回归。 +- P9.0R3 已接受的动作节奏、反馈语言和 P9.0R8 前门不可退化。 -Owns root `AGENTS.md` alignment and `docs/product/`. It defines the complete -loop, recruitment/dismissal/death rules, horizontal formation grammar, -deterministic auto-battle, professions as a fixed catalog schema, economy, -shop refresh, pacing, run structure, difficulty, failure, save and postgame. -It must not enumerate the content catalog owned by the content workstream. +## 架构约束 -### ABG-V2-CONTENT-MATRIX +- 不允许多个并行开发者同时修改 `runtime/main.gd`。 +- 先建立行为等价的运行时拆分边界，再并行实现 domain、UI 与 tests。 +- 领域状态和结果计算不得依赖 Control 节点树或渲染帧率。 +- UI 只通过声明的快照/事件读取领域结果，不复制组合判断。 +- 旧 P9.0R1-R10 测试必须继续通过。 +- 合并 worker 是唯一可修改共享入口和共享 runner 的实现角色。 -Owns `docs/content/`. It replaces the fixed-character bible with generated -recruit specifications and authors the complete profession, trait, equipment, -artifact, enemy, boss and region matrices. It must consume the schemas in this -spec and must not edit product, presentation, prototype or taskgraph files. +## 专家与实现波次 -### ABG-V2-UX-SCREENS +1. 游戏策划、UX/视觉设计、技术架构并行输出三份互斥规格。 +2. PMO 汇合三份规格，消除矛盾并冻结实现合同。 +3. Foundation worker 行为等价拆分共享运行时。 +4. Domain、UI、Tests 三个 worker 在互斥路径中并行实现。 +5. Integration worker 接入共享入口，跑完整回归、矩阵、触摸和正常渲染。 +6. Fresh independent reviewer 固定 revision 只读验收。 +7. Reviewer PASS 后才导出并发布新的私有 H5。 -Owns `docs/presentation/SCREEN_AND_STATE_MAP.md` and -`docs/presentation/UI_AND_VISUAL_SYSTEM.md`. It defines the complete landscape -navigation, shop, recruit, equipment, party-line, battle, result, failure, -settings, accessibility, localization and endgame states. Exact system values -that are still being finalized must use named binding placeholders rather than -inventing a conflicting authority. +## 验收 -### ABG-V2-ART-DIRECTION +- 六个无序双人组合、两种适配和所有声明后续选择有确定性矩阵覆盖。 +- 每个组合完成五战和首局结果，不进入未实现页面，也不使用虚假结果。 +- 至少两个公会/路线/成长选择随队伍状态改变价值；没有单一支配答案。 +- 真实 `Input.parse_input_event` ScreenTouch/ScreenDrag 覆盖完整首局，fallback/direct 为零。 +- 小/参考/大横屏、zh_CN/EN、130%、降低动态、静音和低功耗均有代表状态。 +- 全套 P9.0R1-R10 与 P9.1 测试通过，终态 marker 唯一，严格运行时诊断为零。 +- 正常渲染能看出队列、角色行为、伤害/治疗、选择原因和结果变化。 +- 固定源 reviewer PASS 后才生成并发布远程 H5。 -Owns `docs/presentation/ART_ASSET_AND_ANIMATION_CATALOG.md`. It replaces fixed -character, relationship and lane-grid assets with a modular generated-recruit -system, 42 profession kits, one horizontal battle line, eight region families, -116 enemy identities, equipment readability, animation states and VFX rules. -It specifies production; it does not generate final assets. +## 明确冻结 -### ABG-V2-AUDIO-DIRECTION +- 第二章节及以后、完整 42 职业/320 装备/100 敌人内容生产。 +- 正式 Market、持久装备背包、权威 XP 3-10、晋升、伤亡恢复和 save migration。 +- 权威完整战斗公式、难度、经济、离线进度与长期元进度。 +- 新的付费图片/音频生成、批量最终美术与音频生产。 +- iOS/Android 打包、TestFlight、设备安装、签名、商店、评审与发布。 +- 复制参考商业游戏的名称、文案、数值、布局、代码或资产。 -Owns `docs/presentation/AUDIO_MUSIC_AND_VOICE_CATALOG.md`. It removes the -obsolete 30-character voice burden and specifies readable shop, formation, -auto-battle, casualty, reward, region, UI, accessibility and lifecycle audio. -It specifies production; it does not generate or purchase audio. +## 人工闸 -### ABG-V2-DESIGN-VALIDATION +- Owner 判断完整首局是否看得懂、选择是否有意义、是否愿意继续。 +- Owner 判断视觉和听感；自动测试/截图不得替代。 +- 物理设备、打包和发布保持开放。 -Owns `docs/reviews/DESIGN_VALIDATION_CONTRACT.md` and -`tools/validate_predevelopment_design.py`. It defines deterministic count, -cross-reference, forbidden-concept, schema and coverage checks for the merged -design. It may not change product, content or presentation authority. +## 开放问题 -## Shared Vocabulary - -| Term | Meaning | -|---|---| -| Recruit | One generated run participant, not an authored story protagonist | -| Profession | A fixed reusable combat ruleset referenced by many recruits | -| Trait | One short mechanical modifier with a visible benefit and cost | -| Party line | The ordered simultaneous deployment from front to rear | -| Run | One shop/battle/reward sequence with cumulative recruits and losses | -| Meta progression | Persistent unlock breadth; it may not replace run decisions | -| Artifact | A run-defining shared rule modifier, distinct from unit equipment | - -## Acceptance Criteria - -- No product authority still requires fixed named protagonists, relationships, - a six-unit roster, a three-lane grid or Directive/Anchor/Cache battle rules. -- Recruitment, dismissal, permanent loss, replacement, price pressure, - deployment capacity and cumulative recruitment are explicit and testable. -- Party order changes targeting, protection, reach, healing, movement and - replacement behavior in immediately readable ways. -- Every profession and content row changes a player decision; no name-only or - number-only filler. -- Economy and pacing include five profiles, decision timestamps, solvency and - anti-dominance checks. -- Save/failure rules cover migration, corruption, retry, duplicate claim, - interruption, rollback, clock rollback and bounded offline behavior. -- Cross-document terminology, IDs, counts and formulas are consistent. -- Every player-facing page and key state has a landscape composition, primary - action, readable consequence, empty/error/confirmation behavior and input, - localization and accessibility coverage. -- Art, animation, VFX, music, ambience, SFX and voice scope are production- - countable and match generated recruits rather than fixed protagonists. -- A deterministic design validator checks counts, IDs, cross-references, - forbidden retired concepts and presentation coverage after integration. -- A visible independent reviewer audits the integrated fixed revision before - any HTML prototype or runtime development begins. - -## Excluded Scope - -- Runtime code, engine selection, packages, signing, store or release work. -- Final or bulk production art/audio and paid media generation. -- Copying reference names, prose, formulas, layouts, source or commercial - assets. -- HTML/gallery implementation in this batch; it begins only after the six - design authorities are integrated, independently reviewed and owner-approved. - -## Human Gates - -- Owner accepts or revises the integrated complete-design direction. -- Owner later reviews every landscape UI page and representative art state. -- Runtime, media production and release remain separately authorized. - -## Open Questions - -None blocking this design batch. Exact simultaneous capacity within the 10-12 -target is selected by the five-profile simulation, not guessed from the -reference demo's observed `x/10` display. +无 diff --git a/.agent-taskgraph/staffing/P9_1_TEAM.md b/.agent-taskgraph/staffing/P9_1_TEAM.md new file mode 100644 index 00000000..2dc8cfbc --- /dev/null +++ b/.agent-taskgraph/staffing/P9_1_TEAM.md @@ -0,0 +1,24 @@ +# P9.1 Team Plan + +## Wave 1: specifications + +- Game designer: complete first-session decisions for every starting pair. +- UX/visual designer: exact player-facing screens, states, copy and presentation. +- Runtime architect: safe extraction and non-overlapping development boundaries. + +These visible HAPI sessions may run concurrently because their writes are disjoint. + +## Wave 2: foundation + +One developer extracts shared runtime state/APIs while preserving P9.0R10 behavior. + +## Wave 3: implementation + +- Domain developer: state and transactions only. +- UI developer: real Godot presentation only. +- Test developer: tests, matrices and capture harnesses only. + +## Wave 4: integration and review + +One integration developer owns shared entry files and full regression. One fresh visible +reviewer then audits the immutable candidate read-only. H5 publishing follows only on PASS. diff --git a/AGENTS.md b/AGENTS.md index 8354c4a6..8549fcf6 100644 --- a/AGENTS.md +++ b/AGENTS.md @@ -18,10 +18,13 @@ individual-growth H5 baseline. P9.0R8 is the preserved front-door H5 baseline. P9.0R9 is a preserved first-Company H5 baseline whose separate front/rear selection was rejected by Owner review. P9.0R9A is the preserved automatic- formation baseline: the player chooses any two of four Recruits and the system -orders them. P9.0R10 is the only active runtime slice: carry all six legal pairs -through truthful first-Victory diagnosis, one safety-versus-tempo equipment -decision, a changed second battle and its exact result. It stops at the second -result and continues the fast-H5-before-formal-candidate cadence. +orders them. P9.0R10 is the preserved six-pair compatibility baseline. P9.1 is +the active runtime milestone: carry every legal starting pair through the +complete five-battle first-session shape, while first extracting modular runtime +boundaries so parallel developers do not concurrently edit `runtime/main.gd`. +P9.1 begins with a specification-only expert wave; implementation opens only +after game-design, UX/visual and technical-architecture outputs are reconciled +into one frozen contract. Contract, difficulty, authoritative XP/levels three through ten, promotion, formal Market/permanent equipment inventory, Party order, casualty/recovery, authoritative combat, persistence, chapters, device packaging, stores and diff --git a/CURRENT_CHECKPOINT.md b/CURRENT_CHECKPOINT.md index 501c8b24..a309909d 100644 --- a/CURRENT_CHECKPOINT.md +++ b/CURRENT_CHECKPOINT.md @@ -1,8 +1,9 @@ # Current Checkpoint -Active Goal: `docs/goals/PHASE9_0R10_PAIR_ADAPTATION_GOAL.md` -Active contract: `docs/runtime/PHASE9_0R10_PAIR_ADAPTATION_CONTRACT.md` +Active Goal: `docs/goals/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_GOAL.md` +Active contract: `.agent-taskgraph/spec.md` revision `p9.1-spec-r1` until expert integration freezes `docs/runtime/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_CONTRACT.md` Active ledger: `docs/reviews/PHASE9_GODOT_VERTICAL_SLICE_LEDGER.md` +P9.1 TaskGraph batch: `abg-p9-1-continuous-development-01` P9.0R5 baseline: `6ae33971451912b3cc87c8ffe3b8831fa67c967c` P9.0R5 product preview source: `073ff690ea5a27cb47ee1b563c3eebd6cde9e788` P9.0R4 baseline: `232e3a0a45ec2437ddd528502d487d5264f1c6ae` @@ -30,7 +31,16 @@ P9.0R9A Gallery item: `VL8pU8RUhXNn` P9.0R10 baseline: `d8b359a12f5641257c549894a8114245fa987962` P9.0R10 product preview source: `41204acafa911a505f5b24a493aa032983e706d1` P9.0R10 Gallery item: `gHS8jEZdGa2x` -Stage: `P9_0R10_PAIR_ADAPTATION_H5_PREVIEW_READY` +Stage: `P9_1_EXPERT_SPEC_WAVE_PENDING_DISPATCH` + +P9.1 owns the next bounded runtime milestone. It must preserve P9.0R10 and carry +all six legal starting pairs through the existing five-battle first-session shape +with truthful Guild response, route, growth and final-result behavior. The first +wave is specification-only: game design, UX/visual design and runtime architecture +work in disjoint files, then PMO freezes one implementation contract. No runtime +implementation may begin before that integration gate. Chapters, bulk content, +authoritative full combat/economy/save, paid media, packaging, devices, stores and +release remain frozen. P9.0R10 owns the next compatibility boundary after automatic formation. All six legal two-Recruit pairs must receive a truthful first-Victory diagnosis, choose diff --git a/docs/goals/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_GOAL.md b/docs/goals/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_GOAL.md new file mode 100644 index 00000000..ef2dfc22 --- /dev/null +++ b/docs/goals/PHASE9_1_ARBITRARY_PARTY_FIRST_SESSION_GOAL.md @@ -0,0 +1,53 @@ +# Goal: Phase 9.1 Arbitrary-Party Complete First Session + +> Task family: `ABG-P9.1` +> Started: 2026-08-13 +> Baseline closure: `f0b7d7b0234d8ad304f7b9d39fa99ded1dbfc27d` +> Accepted product source: `41204acafa911a505f5b24a493aa032983e706d1` +> TaskGraph batch: `abg-p9-1-continuous-development-01` +> Legal terminal: **Aetherbound Guild P9.1 candidate ready** or **Aetherbound Guild P9.1 redesign required: ** + +## Objective + +Turn the six-pair P9.0R10 compatibility preview into one complete, understandable first +session for every legal starting pair, while extracting modular runtime boundaries that let +domain, UI and test developers work without concurrent edits to `runtime/main.gd`. + +## Required Player Arc + +`choose any two -> auto formation -> battle 1 -> diagnosis -> adaptation -> battle 2 ->` +`Guild response -> battle 3 -> route -> battle 4 -> individual growth -> battle 5 -> result` + +Each choice and result must be based on the actual party, equipment and prior decisions. + +## Delivery Stages + +1. Parallel game-design, UX/visual and runtime-architecture specifications. +2. PMO integration into one implementation contract. +3. Behavior-equivalent runtime extraction. +4. Parallel domain, UI and test implementation with disjoint writes. +5. Integration, full regression and source-bound runtime evidence. +6. Fresh independent review. +7. New private remotely playable H5 after reviewer PASS. +8. Owner play judgment. + +## Acceptance + +- All six starting pairs complete the five-battle session without false or unsupported states. +- At least two rational responses change value with party state and previous decisions. +- Results name action, cause, changed state and next action in readable Chinese. +- Existing accepted formation, combat movement, feedback and front-door behavior do not regress. +- Tests cover pair paths, negative transaction boundaries, real parsed touch, layouts, + accessibility variants and strict diagnostic cleanup. +- Repository is clean/upstream at submission; every worker commit is focused and pushed. +- Reviewer PASS precedes H5 publishing; Owner understanding/fun remains open until played. + +## Frozen + +No chapter expansion, bulk content, authoritative full combat/economy/save systems, paid media, +mobile packaging, TestFlight, device actions, store actions or release work. + +## Current Gate + +The first three expert tasks must produce mutually consistent specifications. Runtime code is +not authorized until their commits are integrated and the implementation contract is frozen. diff --git a/docs/reviews/PHASE9_GODOT_VERTICAL_SLICE_LEDGER.md b/docs/reviews/PHASE9_GODOT_VERTICAL_SLICE_LEDGER.md index 2248ab4a..870228e8 100644 --- a/docs/reviews/PHASE9_GODOT_VERTICAL_SLICE_LEDGER.md +++ b/docs/reviews/PHASE9_GODOT_VERTICAL_SLICE_LEDGER.md @@ -40,3 +40,4 @@ separate. | 2026-08-13 | product `32b4fdb`; Gallery `VL8pU8RUhXNn` | `P9_0R9A_AUTO_FORMATION_H5_PREVIEW_READY` | Unified selection shows all four Recruits; the player selects/removes any two and never chooses a slot. All six unordered pairs in both click orders produce the same system field order and first Victory, including Hearthguard/Wayrunner and Farstring/Kindhand. Capacity, unified Revise, once-only commit and cleanup pass; parsed ScreenTouch is `touch=11 fallback=0 direct=0`. R1-R8 standard-pair gameplay, full three-battle touch, animation and audio regressions pass after restoring the frozen R3 qualifier. Ten `844x390` Chinese/English-130% normal-render states have 44px controls and clean final diagnostics. The runtime-only PCK excludes tests/evidence/build; nine payloads verify against `export.sha256` SHA-256 `20785617bccb3ee646f41cbd3c576757e626ccce9cd8c2c63cd2845be4fb99c2`. Remote MIME/length and byte-identical PCK pass. Owner understanding/fun/visual/listening and later-cycle support for exceptional pairs remain open. | | 2026-08-13 | Owner continuation from clean/upstream `d8b359a` | `P9_0R10_PAIR_ADAPTATION_ACTIVE` | R10 carries all six R9A pairs through first-Victory diagnosis, the existing unselected `EQ-002` safety versus `EQ-061` tempo decision, a pair-aware second battle and exact result. It repairs the exceptional second-position Wayrunner fallthrough without changing accepted standard-pair values. R5-R7 later systems, authoritative combat, save, Market, content, packaging and release remain frozen. | | 2026-08-13 | product `41204ac`; Gallery `gHS8jEZdGa2x` | `P9_0R10_PAIR_ADAPTATION_H5_PREVIEW_READY` | All six pairs and two equipment choices pass 12 exact paths: safety HP `90/100` at `7.05s`, tempo HP `70/82` at `6.00s`, Coin 20, and actual Farstring ranged, Kindhand healing or second-position Wayrunner counter behavior. Parsed touch is `touch=5 fallback=0 direct=0`; 22 serial R1-R10 tests pass with strict markers zero. Eight `844x390` states pass 44px bounds after a fresh repair for English-130% result overflow. Browser inspection repaired the stale Title front/rear claim before final export. Runtime-only PCK inventory is 227 files with forbidden paths zero; nine-file manifest SHA-256 is `7aef140b95d4c5d5bcc5c5815f4ae02894ee49367a6fbeb87dc07915d258f859`, and remote PCK equals local. Gallery Canvas reaches the corrected Chinese Title with browser diagnostics zero after roughly four-minute cold start. Human comprehension/fun/visual/listening, exceptional-pair post-battle-two support, persistence, device, packaging, stores and release remain open. | +| 2026-08-13 | Owner P9.1 TaskGraph authorization from closure `f0b7d7b` | `P9_1_EXPERT_SPEC_WAVE_PENDING_DISPATCH` | Owner authorizes visible parallel game-design, UX/visual, technical-architecture and later developer sessions. Batch `abg-p9-1-continuous-development-01` first freezes three disjoint specifications, then performs exclusive runtime extraction before parallel domain/UI/test implementation, integration, fresh review and H5 publishing. P9.0R10 remains the accepted product source. Runtime code is still frozen at this gate; HAPI control-plane verification and exact identity ACK are required before any expert task becomes active. Chapters, bulk content, paid media, packaging, devices, stores and release remain frozen. |