* feat: improve spawn error handling and reporting across full stack
- Return error result instead of throwing in apiMachine spawn handler
- Add lastSpawnError field to RunnerState for persistent error tracking
- Add error awaiter system for early process exit/error detection before webhook
- Build detailed webhook failure messages with exit code, signal, and stderr tail
- Report spawn outcomes to hub via runner state updates
- Handle more spawn result types in rpcGateway with better error messages
- Display runner last spawn error in web UI (NewSession & SpawnSession)
- Extract shared formatRunnerSpawnError utility to avoid duplication
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(cli): narrow spawnResult type check to fix TS2339 error
Use `type === 'error'` instead of `type !== 'success'` to properly
narrow the discriminated union, allowing TypeScript to infer errorMessage.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat: Add Claude Code Agent Teams support
- Add TeamState schemas and types for team collaboration
- Extract team state from TeamCreate, SendMessage, Task tools
- Add database migration V3→V4 for team_state storage
- Add TeamPanel component to display team members, tasks, messages
- Add team tool icons and presentation rules
- Support vite proxy configuration via VITE_HUB_PROXY env var
via [HAPI](https://hapi.run)
Co-Authored-By: HAPI <noreply@hapi.run>
* fix: Add timestamp protection for team_state updates
Prevent old messages from overwriting newer team state by checking
team_state_updated_at before updating.
via [HAPI](https://hapi.run)
Co-Authored-By: HAPI <noreply@hapi.run>
* fix: Extract team tasks from Task/TaskCreate/TaskUpdate tools
- Enhance processTaskToolWithTeam to also generate task entries from
the Task tool's description field when spawning teammates
- Add processTaskCreate handler for TaskCreate tool calls
- Add processTaskUpdate handler for TaskUpdate tool calls
- Register both new tools in the extraction switch statement
This fixes the gap where the Tasks section in TeamPanel could never
populate because team task data was not being extracted from the
message stream.
via [HAPI](https://hapi.run)
Co-Authored-By: HAPI <noreply@hapi.run>
* fix: Skip orphan TaskUpdate without title to prevent schema validation failure
When TaskUpdate arrives before TaskCreate (message ordering), skip inserting
incomplete tasks that lack required title field, preventing entire teamState
from being dropped by schema validation.
via [HAPI](https://hapi.run)
Co-Authored-By: HAPI <noreply@hapi.run>
* test: Add unit tests for orphan TaskUpdate handling
Verify that applyTeamStateDelta correctly skips inserting tasks without
title field (orphan TaskUpdate) while still allowing normal task creation
and updates to existing tasks.
via [HAPI](https://hapi.run)
Co-Authored-By: HAPI <noreply@hapi.run>
---------
Co-authored-by: tfq <tfq@gmail.com>
Co-authored-by: HAPI <noreply@hapi.run>
Add project-level slash command discovery with recursive nested command scanning, pass workingDirectory through slash-command handlers, and align hub/web source unions to include project commands.
* feat(cursor): add support for Cursor Agent CLI integration
- Introduced new command `hapi cursor` to start Cursor Agent sessions.
- Added functionality for resuming sessions and managing permission modes.
- Updated documentation to include Cursor Agent usage and installation instructions.
- Enhanced existing codebase to accommodate Cursor as a recognized agent flavor.
- Implemented local and remote session handling for Cursor Agent.
This update expands HAPI's capabilities by integrating support for the Cursor Agent, allowing users to leverage its features alongside existing agents.
* Remove TODO.md file as it is no longer needed following the integration of Cursor Agent CLI support. This cleanup helps streamline project documentation and reflects the completion of the associated tasks.
* feat(cursor): implement remote mode and fix --hapi-starting-mode
- Consume --hapi-starting-mode in cursor command (do not forward to agent)
- Implement cursorRemoteLauncher: spawn agent -p with stream-json, --trust
- Add cursorEventConverter for NDJSON parsing (system/assistant/tool_call/result)
- Multi-turn via --resume session_id
- Update docs: cursor supports both local and remote modes
Made-with: Cursor
* fix: type error
* fix(cursor): address PR review - model UI, sessionId metadata, duplicate flags
- HappyComposer: use isClaudeFlavor for model mode (cursor has no model modes)
- cursorLocalLauncher: call onSessionFound for resume so cursorSessionId in metadata
- cursorCommand: do not forward parsed flags to cursorArgs (avoid duplicates)
Made-with: Cursor
Refactor platform detection and error reporting logic into dedicated,
testable functions. This improves code organization and enables better
error handling for unsupported platforms and missing platform packages.
- Add SUPPORTED_PLATFORMS constant with platform labels
- Extract getPlatformKey() utility function
- Extract isSupportedPlatform() validation function
- Extract reportUnsupportedPlatform() error reporting function
- Extract reportMissingPlatformPackage() error reporting function
- Update main() to use new functions for clearer error handling
Extract platform detection and error reporting logic into reusable functions
to improve testability and reduce code duplication in the binary launcher.
Adds comprehensive test coverage for platform validation and error messages.
Replace hardcoded 'claude' string with getDefaultClaudeCodePath() to
ensure the correct Claude Code executable path is resolved, consistent
with claudeLocal.ts. This respects HAPI_CLAUDE_PATH env var and finds
the global claude installation path properly.
* feat(cli): support slash commands from installed plugins
Allows HAPI to automatically scan and load slash commands from installed Claude plugins (e.g. superpowers). Commands are indexed with their plugin namespace prefix.
* fix(web): add 'plugin' to SlashCommand source type
This fixes the typecheck error identified by the PR review bot.
* fix(web): update Suggestion type to include 'plugin' source
* fix(slashCommands): 修复 scoped 插件名称解析及多版本安装选择逻辑
- 使用 lastIndexOf('@') 替代 split('@')[0] 以正确解析带 scope 的插件名
- 按 lastUpdated 降序排序选择最新的安装实例
- 增加 installations 空数组边界检查
On Windows, spawning Claude with shell: true causes the process to
exit immediately with code 1. This happens because shell: true invokes
cmd.exe /c claude <args>, and cmd.exe's PATH resolution differs from
direct process creation, leading to environment inconsistencies.
This fix:
- Adds findWindowsClaudePath() to locate claude.exe absolute path
- Changes spawn to use absolute path with shell: false on Windows
- Maintains Unix behavior (command name works fine with shell: false)
- Adds HAPI_CLAUDE_PATH env var for user override
Tested on Windows 11 with Claude Code v2.1.29 and hapi v0.15.0.
via [HAPI](https://hapi.run)
Co-authored-by: HAPI <noreply@hapi.run>
