mirror of
https://github.com/wu736139669/hapi.git
synced 2026-08-05 06:24:37 +00:00
55d1bbb7bd946ecf21f5625aeabe98cb3f90e1a7
816
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
|
55d1bbb7bd |
feat(cursor): invisible sync-on-open migrator from legacy stream-json to ACP (#844)
* feat(cursor): invisible sync-on-open migrator from legacy stream-json to ACP Closes #824 When the operator reopens a legacy stream-json Cursor session in HAPI, the hub now transparently transplants its `~/.cursor/chats/<wsh>/<uuid>/store.db` into `~/.cursor/acp-sessions/<uuid>/`, verifies it loads via `agent acp`, flips `metadata.cursorSessionProtocol = 'acp'`, and removes the legacy source - all before `resumeSession` returns. Subsequent opens are pure ACP. The primary justification is safety, not feature parity. #784 (`cursor-agent` fabricates `Questions skipped by the user` responses in legacy stream-json mode) still fires regularly in dogfood despite #801's mitigation: the agent ships destructive side effects against fabricated consent. Migration to ACP closes the protocol-level door because the `AskQuestion` tool does not exist on the ACP side, so there is nothing to fabricate. working. That tradeoff was reasonable at the time. The accumulated #784 evidence makes legacy sessions actively unsafe; this PR makes the upgrade path invisible enough that users stop avoiding it. A pre-PR spike established that legacy and ACP `store.db` files use the identical SQLite schema; only the directory layout differs. The migrator therefore: 1. Sanity-checks the source store and pre-flips state (`session.active`, `lifecycleState`, on-disk presence, target collision) 2. Optionally archives a stale-running row (`forceArchiveRunning: true` is the default for the auto-migrate path because the caller already verified `session.active === false`) 3. Atomically creates `~/.cursor/acp-sessions/<uuid>/` with mode `0o700` 4. Copies `store.db` and chmods to `0o600` (multi-user-host hardening) 5. Writes a minimal `meta.json` sidecar (`schemaVersion`, `cwd`, optional `title`) with mode `0o600` 6. Spawns `agent acp` under HAPI_HOME isolation and verifies the session loads via `session/load`. On long histories the verify also drives a trivial single-turn prompt; on short ones load-only is enough 7. Flips `cursorSessionProtocol = 'acp'` AND clears the `cursorMigrationState` banner flag in a SINGLE metadata write 8. Removes the legacy source store (only after verify succeeded and the protocol flip committed). The legacy `~/.cursor/chats` parent dir is left as-is Every failure leaves the legacy state intact. No `rm` fires without a verify success AND a committed protocol flip. The transplant takes 15-20s on long histories (copy a multi-hundred-MB store, spawn `agent acp`, replay thousands of notifications, tear down the probe). Without a progress indicator the wait reads as "broken" to a fresh reviewer. A minimal banner ships alongside the migrator: - Hub sets `metadata.cursorMigrationState = 'in_progress'` BEFORE the long-running transplant. The session-cache refresh emits the existing `session-updated` SSE event (no new event type), so the web client picks it up in milliseconds. No client-side polling needed. - Hub clears the flag in the SAME metadata write that flips `cursorSessionProtocol` to `'acp'` on success, so the banner disappears in the same render tick the chat re-renders as ACP - no flicker window. - Hub clears the flag explicitly in the auto-migrate helper's `finally` on failure/exception, so the banner never gets stuck if migration falls back to the legacy launcher. - Web renders an accessible (role=status, aria-live=polite) banner with an indeterminate spinner. Deliberately no fake percentage - we do not have phase data and a fake progress bar would lie. This PR is intentionally sequenced AFTER swear01's three ACP mop-up PRs (merged today as |
||
|
SSU-WEI HUANGandGitHub
|
ddf3a5545b | fix(web): add missing i18n keys for session.inactive banner (#851) | ||
|
|
cad58cfa0b |
fix(opencode): use ACP-reported reasoning effort options (#853)
* fix(opencode): use ACP-reported reasoning effort options Expose thought_level options from OpenCode ACP to the web UI via RPC/API instead of hardcoded presets, and validate effort values before setConfigOption. Fixes #852 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(opencode): sync hub effort after coerced setConfigOption When resolveThoughtLevelEffort falls back to a different supported value, roll back session state after a successful ACP update so keepalive and the web UI do not keep advertising the rejected effort. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
3473a88d67 |
feat(web): auto-return to chat when remote terminal exits (#857)
When the shell exited inside the remote terminal view, the page kept
showing a banner ("Terminal exited with code 0.") and left the user
stranded with no obvious next step. On mobile this is awkward, and it
does not match the muscle memory from native terminal emulators where
typing `exit` closes the tab/window.
Schedule a goBack() shortly after `terminal:exit` fires so the user
briefly sees the exit info, then returns to the session chat (same
destination as the existing back arrow via useAppGoBack).
The auto-close timer is cleared on unmount, on sessionId change, and
when the socket reconnects after a transient drop so a stale exit
event cannot navigate away from a freshly reconnected terminal.
Closes #856
Co-authored-by: Cursor <cursoragent@cursor.com>
|
||
|
hangerandGitHub
|
e2625b8b47 |
feat: add Fable model presets for Claude sessions (#860)
* feat: add Fable model presets for Claude sessions Claude Code 2.x accepts the fable / fable[1m] model aliases for Fable 5. hapi passes the model string through verbatim, so adding the presets to CLAUDE_MODEL_LABELS surfaces them in the new-session and composer model pickers, labels, and the 1M context-window heuristic. * test: update modelOptions full-list assertions for Fable presets Addresses review feedback on #860: getModelOptionsForFlavor appends every Claude preset, so the two complete-array expectations must include the new fable entries. |
||
|
weishu
|
1f92a31b12 | Release version 0.20.1 | ||
|
|
393cd7bfbb |
feat(web): scratchlist v1.1 — composer-toggle drawer + reusable FUE primitive (#798)
* feat(web): scratchlist v1.1 — composer-toggle drawer + reusable FUE primitive The v1 always-visible amber band proved too heavy for what's a 20% feature in a typical session. v1.1 dials it back to a composer-toggle that opens an on-demand drawer, paired with a reusable FUE (First-User Experience) primitive so existing operators get a subtle pulsing dot + on-click explainer the first time they see the toggle. UX changes: - Notepad icon in the composer toolbar (next to schedule-send) toggles scratchlist mode. Drawer renders only while mode is on. - Composer's send button repaints amber and reads "Send to scratchlist" while the mode is sticky; submit routes adds into the scratchlist instead of the chat. Click the icon again to leave. - Small entry-counter badge appears on the toggle when entries exist; empty-state shows just the icon (no zero-state guilt UI). New reusable FUE primitive: - web/src/lib/use-fue.ts: state machine (unseen → engaging → acknowledged) with localStorage persistence, namespaced under hapi.fue.v1.<featureId> so it can't collide with any future upstream onboarding flow. - web/src/components/Fue.tsx: <FueDot> (small pulsing badge) and <FueCallout> (portal-rendered popover with title/body + "Got it" affirmative-action dismiss). No auto-timeout — reading speed varies and silent disappearance undercuts user trust. - AGENTS.md adds a "Adding new web features — consider an FUE" section so future contributors discover the primitive. Refactors: - ScratchlistPanel.tsx: split rendering into <ScratchlistInventory> (presentational list) and <ScratchlistDrawer> (composer-controlled drawer with hint copy). Original <ScratchlistPanel> kept exported for the existing fixture-based tests. - SessionChat.tsx: scratchlist state lifted into useScratchlist hook so the composer-toolbar counter and the drawer share one source of truth. onSend wrapped to route through scratchlist.add when mode is on. Tests: - 9 useFue hook tests (initial state, engage idempotency, no auto-acknowledge, dismiss, featureId switching, post-acknowledged engage no-op, resetFue helper). - 5 placement helper tests (above/below switching, viewport edge clamping, visualViewport offset support). - All 21 existing scratchlist lib tests + 14 ScratchlistPanel tests continue to pass. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): prevent cross-session leak in useScratchlist hook Per upstream review on PR #798 (github-actions[bot] [Major]): > useScratchlist persists current `entries` whenever `sessionId` > changes. On A -> B navigation, React first commits with B's id > and A's entries; after paint, this persist effect can write A's > entries to hapi.scratchlist.v1.B before the rehydrate effect > loads B. The previous keyed panel existed specifically to avoid > this race. Lifting state out of the v1 panel (which sidestepped the race via key={props.session.id} forced remount) re-introduced this same data- loss window. The composer-controlled drawer in v1.1 cannot remount on session change because its parent SessionChat doesn't either. Fix: keep the loaded sessionId in state alongside the entries so they swap atomically, and persist against the LOADED sessionId rather than the prop. After A->B, the loaded sessionId is still A until rehydrate runs, so a spurious persist re-writes A's storage with A's entries - a no-op instead of a corruption. Tests: - New use-scratchlist.test.ts with 6 tests: - hydrates from localStorage on mount - add() persists to current session's storage only - rerender to a new session preserves the new session's existing entries - after switching, add() targets the new session - regression test that spies on Storage.prototype.setItem and asserts the rerender lifecycle never produces a (B-key, A-entries) write - remove()/move() target the loaded sessionId - The setItem-spy test correctly fails against the buggy code (verified by temporarily reverting the fix) and passes with the fix in place. - Full web suite: 88 files, 756 tests, all green. Typecheck clean. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): route attachment/scheduled submits to chat instead of dropping them Per upstream review on PR #798 (github-actions[bot] [Major]): > Prevent scratchlist mode from dropping attachments — in scratchlist > mode the wrapper returns success after adding only `text`, while > HappyComposer still treats composer attachments as sendable input. > A text+attachment submit therefore routes through this branch, > stores only the text, and silently discards the attachment instead > of sending or preserving it. Same hazard applies to scheduledAt: scratchlist entries are pure-text notes - they can't represent attachments or schedule metadata - so any submit carrying either MUST fall through to props.onSend (chat) even when the scratchlist toggle is on. Otherwise the wrapper short-circuits to scratchlist.add(text), reports success to the composer, and the composer dutifully clears attachments + schedule that the user just queued. Fix: extracted the routing rule into shouldRouteToScratchlist(mode, attachments, scheduledAt) - returns true only when mode is on AND the payload is pure text. onSendForComposer uses it. Tests: - 5 new shouldRouteToScratchlist unit tests (mode off, mode on + text-only, mode on + attachments, mode on + schedule, mode on + both) - All in web/src/components/SessionChat.test.ts (13 tests total now) - Full web suite: 88 files, 761 tests, all green. Typecheck clean. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): clear pendingSchedule when scratchlist-mode submission falls back to chat Per upstream review on PR #798 (github-actions[bot] [Major]): > The follow-up change correctly falls through to props.onSend when > scratchlist mode is on but scheduledAt is present, yet the > accepted-send cleanup still checks only !scratchlistMode. That > means a scheduled chat send made while the amber scratchlist UI is > active is accepted, but pendingSchedule stays set, so the next > normal send can accidentally reuse the same schedule. Fix: handleSend now gates the cleanup branch on the actual route taken (routedToScratchlist) rather than the scratchlist UI state. Reuses the same shouldRouteToScratchlist helper so route + cleanup share a single source of truth. Tests: - 2 new tests in SessionChat.test.ts that pin the decision matrix handleSend depends on: - 'cleanup gate: scheduled chat send while scratchlist toggle is on still clears schedule' - 'cleanup gate: pure-text scratchlist add does NOT clear schedule' - Full web suite: 88 files, 763 tests, all green. Typecheck clean. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): UnifiedButton must reflect actual routing, not raw scratchlist toggle Per upstream review on PR #798 (github-actions[bot] [Major]): > Send button advertises scratchlist routing even when the submit > will go to chat — shouldRouteToScratchlist correctly falls back > to normal chat for attachments or scheduledAt, but UnifiedButton > still turns amber and labels the action as "Send to scratchlist" > whenever scratchlistMode is true. A scheduled send or attachment > send made in that state will be submitted to chat while the UI > says it is being stashed, which can send content to the agent > unexpectedly. Fix: - UnifiedButton's prop renamed `scratchlistMode` -> `routesToScratchlist` to make the contract explicit: "this submit really will go to the scratchlist", not "the scratchlist toggle is on". - The call site computes `routesToScratchlist` from `scratchlistMode && !hasAttachments && pendingSchedule == null`, mirroring SessionChat's shouldRouteToScratchlist exactly. The button is now amber + "Send to scratchlist" only when the actual send path will hit scratchlist; attachments / pending schedule force a chat- style render that matches the real routing. - UnifiedButton exported so it can be unit-tested directly. Tests: - 3 new render tests in ComposerButtons.test.tsx covering: - routesToScratchlist=true → amber + "Send to scratchlist" - routesToScratchlist=false → black + "Send" (the regression case) - omitted prop → defaults to chat-style render - Full web suite: 89 files, 766 tests, all green. Typecheck clean. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): exit scratchlist mode when promoting an entry to the composer Per upstream review on PR #798 (HAPI Bot, follow-up after b256fe5): > Found one major issue: promoting a scratchlist item to the composer > keeps scratchlist mode enabled, so the next send re-adds it to the > scratchlist instead of sending to chat. Promoting an entry to the composer means "I want to send this for real now". With scratchlist mode still on, the next composer submit routes back to scratchlist (per the v1.1 modal-mode contract), so the user's click loop becomes promote -> send -> re-add -> nothing-actually-sent. Fix: ScratchlistDrawerHost now calls onExitScratchlistMode whenever it promotes an entry to the composer. Promote-to-queue does NOT exit the mode (queue path bypasses the wrapper anyway, and the operator may still be capturing related notes). Tests: - Exported ScratchlistDrawerHost so its host-level callbacks can be unit-tested in isolation (previously only ScratchlistDrawer was testable; the wiring was untested). - New SessionChat.exit-mode.test.tsx with 2 tests: - promote-to-composer fires setText AND onExitScratchlistMode - promote-to-queue fires onSend but does NOT exit mode - Full web suite: 90 files, 768 tests, all green. Typecheck clean. Co-authored-by: Cursor <cursoragent@cursor.com> * feat(scratchlist): Ctrl/Cmd+Shift+S toggles scratchlist mode (v1.1 hotkey) The v1 always-visible panel had Ctrl/Cmd+Shift+S to expand the panel and focus the input. v1.1 mounts the drawer only when scratchlistMode is on, so the v1 listener (inside the panel) is dead code: it can't fire while the drawer is unmounted, and the user has no way to open the drawer without clicking the toolbar icon. Re-bind the shortcut at SessionChat scope so it's always alive and toggles the mode. Convention matches sibling globals (Ctrl/Cmd-m cycles agent model). Ctrl/Cmd-Shift-S is unreserved by Chrome / Firefox / Safari (browser Save As is Ctrl-S / Cmd-S, no Shift), so the user's save-page muscle memory keeps working. Modifier requirement (Ctrl/Cmd+Shift) means it can't collide with literal-character typing in any input - no focus suppression needed. The matcher is extracted to a pure helper isScratchlistToggleHotkey so it's unit testable without mounting SessionChat. 6 new tests pin the modifier matrix: - Ctrl+Shift+S (Linux/Windows) -> match - Cmd+Shift+S (macOS) -> match - Cmd/Ctrl+S without Shift -> reject (browser Save reservation) - bare S / Shift+S -> reject (literal typing) - Ctrl+Shift+Alt+S -> reject (avoid OS clashes) - other modifier+key combos -> reject Tooltip + FUE body now mention the hotkey so it's discoverable from the same UI surface that introduces the feature (en + zh-CN). Web suite 90 files / 774 tests, all green. Typecheck clean. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): hotkey skips dialogs / inputs / contentEditable Bot finding on PR #798 (PRRT_kwDOQuQOSc6HGtLn): the window-level Ctrl/Cmd+Shift+S listener fires for every focus target, so the shortcut can toggle scratchlist mode "behind" an open modal (rename session, schedule picker, FUE callout, image preview), making the next composer send route to scratchlist instead of chat. UX bug. Add isScratchlistHotkeyBlockedTarget(target) and gate the listener on it. Block targets: - any descendant of an open [role="dialog"] (Radix UI's DialogContent renders role="dialog"; FueCallout, ScheduleTimePicker, ImagePreview also use role="dialog") - HTMLInputElement (single-line inputs) - HTMLSelectElement - any contentEditable host (with attribute-based fallback for jsdom, which doesn't implement isContentEditable) NOT blocked: - HTMLTextAreaElement (the composer textarea is the expected focus target when the operator presses the hotkey - blocking would defeat the shortcut) - the document body / unfocused targets 8 new unit tests pin the matrix. Function is exported / pure so callers can reuse the same blocked-target rule for future global shortcuts. Suggested fix from the bot applied modulo: - Use !== null on closest() result (explicit boolean for return type) - Add attribute-based contentEditable fallback for jsdom test env * feat(scratchlist): copy-to-clipboard action on each entry Add a per-entry "Copy to clipboard" button between the send-to-queue and delete actions. On click, write the entry text via the shared safeCopyToClipboard helper (which already handles the navigator.clipboard primary path + the execCommand fallback for Safari / non-secure-context edges); on success, briefly flip the icon to a check and the aria-label/title to "Copied!" for 1500ms so the operator gets visual + screen-reader confirmation. Failures (clipboard denied AND execCommand fallback unavailable) silently no-op rather than throw at the click handler. Mirrored across both surfaces: - ScratchlistInventory (used by the v1.1 composer-toggle drawer) - ScratchlistPanel inline list (the v1 always-visible panel) A small useCopiedFeedback() hook owns the "which entry just got copied" state + the 1.5s auto-clear timeout. Pure state machine; the caller wires safeCopyToClipboard separately so the hook itself stays free of jsdom clipboard quirks. Cleared on unmount via the standard ref-tracked timeout pattern, so promote-and-navigate-away can't leak. Locale keys: scratchlist.action.copy / scratchlist.action.copied (en + zh-CN). Three new tests: - v1 panel happy path: writeText called with the entry text, button flips to the "Copied!" label, entry is preserved (copy is non-destructive). - v1 panel failure path: writeText rejects AND execCommand returns false; button stays in "Copy to clipboard" state — no false success. - v1.1 drawer happy path: writeText called, label flips, and crucially no other entry handlers (onSend, onDelete, setText, onExitScratchlistMode) fire — copy is independent of all the other actions. Web suite 90 files / 785 tests, all green. Typecheck clean. * fix(scratchlist): reset all per-session state via keyed wrapper Bot finding on PR #798 (PRRT_kwDOQuQOSc6HHOsa): when the operator navigates between sessions on the same route (/sessions/A -> /sessions/B), React reuses the SessionChat component instance. Effects run AFTER the first paint, so for a single render window the new session is rendered with the previous session's scratchlist entries (useScratchlist's rehydrate-effect) AND drawer-open state (scratchlistMode reset effect). Visual leak; drawer actions targeting stale state. Apply the bot's suggested fix verbatim modulo the type extraction: export function SessionChat(props) { return <SessionChatInner key={props.session.id} {...props} /> } Canonical React idiom for "fully reset state on prop change": the keyed wrapper unmounts and re-mounts the inner component when session.id changes, so every hook (useScratchlist's initial-state factory, useState, useHappyRuntime, ...) starts fresh. This supersedes the now-redundant effect-based reset: - useEffect(() => { setScratchlistMode(false) }, [session.id]) REMOVED useScratchlist's atomic-loaded-sessionId persistence (added on the prior PR round) stays as defense-in-depth for any caller that uses the hook without the keyed-wrapper pattern. Web suite 90 files / 785 tests, all green. Typecheck clean. * fix(web): retain composer text on send failure (closes #776) When the message composer submits and the hub responds with a 4xx/5xx or the fetch fails outright, assistant-ui clears the composer synchronously the moment send is invoked. Without intervention the operator's typed text is destroyed at exactly the moment they most need it preserved. SessionChat additionally clears any pending schedule on accept, so a failed scheduled send was also silently downgrading to immediate on the next attempt. Behaviour: - useSendMessage exposes onError({ sessionId, text, scheduledAt, error }) so the route can hand the input back to the composer. sessionId is the resolved target (post-resolveSessionId), so an inactive-session resume that resolves a new id, kicks off async navigation, then fails the POST restores into the resumed session's composer rather than the old one. - router.tsx stores sendErrors keyed by sessionId. Per-session lookup replaces the clear-on-session-change effect, so errors do not bleed between sessions and a session-scoped failure persists across navigation. - HappyComposer accepts ComposerSendError, restores text via api.composer().setText() once per failure id, and re-establishes any pending schedule via onSchedule({ type: 'absolute', ms: scheduledAt }). It renders a red ring on the composer wrapper and a role="alert" inline message; both clear the moment the operator types or sends. - onError forks on input.attachments. Text-only sends use the composer-restore path (removeOptimisticMessage drops the row so the failed bubble does not duplicate the restored text). Attachment sends keep the legacy failed-bubble UX (status='failed' + in-thread retry button) because the composer-restore path can't reinstate uploaded attachment metadata. retryMessage extracts attachments from the stored optimistic message via getMessageAttachments so failed-bubble retry of an attachment send re-fires with its files. Acceptance (issue #776): - Submit -> 500/502/503/network error -> composer text not cleared - Submit -> 400/401/403 -> composer text not cleared, error inline - Submit -> 2xx -> composer clears as today - Operator can edit retained text and retry without re-typing - Failed scheduled sends restore as scheduled, not as immediate Tests in web/src/hooks/mutations/useSendMessage.test.tsx cover text-only 4xx/5xx/network retention, scheduled-send carry-through, optimistic-row removal on text-only failure, sessionId carry-through under resolveSessionId, attachment failure fallback, and attachment retry preservation. Full web suite passes (705 tests). bun typecheck clean. No SCHEMA_VERSION bump (frontend-only). * fix(test): correct AttachmentMetadata fixture shape + JSX namespace import Two pre-existing test-only typecheck failures surfaced once scratchlist v1.1 was stacked into the driver soup. * SessionChat.test.ts - the attachment() fixture used the legacy schema (kind, sizeBytes) instead of the current AttachmentMetadataSchema (filename, size, path). Updated to match the live shape so the cast is honest. * ComposerButtons.test.tsx - JSX namespace is no longer global under the current TS lib config; switched the helper signature from JSX.Element to React's ReactElement (same runtime, named import). Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): soften panel chrome - drop strong amber fill, keep subtle accent border (#812) Per #812 (and PR 827 from @swear01) the always-visible amber fill on the scratchlist panel was too loud as a scroll element. This swaps the warning *fill* for the chat-user-surface tone and uses neutral text/pills/focus, but keeps the warning *border* as a soft accent so the panel still reads as a different destination from a normal user message. The strong destination signal continues to live on the composer Send button (it goes amber-500 only while scratchlist mode is routing) and the active toggle button - those carry the moment-of-action signal the user actually presses, and ComposerButtons tests + the FUE copy already depend on that behavior, so they're unchanged. Credit to @swear01 (PR 827) for the styling note; this branch absorbs that restyle and supersedes the Settings-toggle approach because v1.1 hides the panel by default behind the composer drawer toggle (no Settings entry needed). Adds a regression-guard test asserting the panel uses the chat-user-surface bg + warning-border (not the warning fill). Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
weishu
|
deb05bb783 | Auto-approve Codex title MCP tool | ||
|
|
fa363c2f6c |
fix(cursor): register cursorSessionId before ACP session/load (#837)
Pre-write resume token into session metadata before awaiting session/load so hub and web see cursorSessionId immediately after resume spawn. Fixes #834 Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
8094b500f3 |
fix(web): hide sidebar fake sessions for Cursor resume/archive (#836)
* fix(web): dedupe sidebar sessions by flavor resume id Wire deduplicateSessionsByAgentId into SessionList and resolve cursor threads via cursorSessionId so resume/archive no longer shows duplicate inactive rows for the same ACP session. Fixes #833 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): use SessionSummary.agentSessionId for sidebar dedup SessionList only receives SessionSummary from the API; native ids like cursorSessionId are already mapped into metadata.agentSessionId by toSessionSummary. Drop resolveAgentSessionIdFromMetadata to fix typecheck. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): hide inactive empty session stubs in sidebar Filter inactive rows with no agentSessionId and no title signal before grouping sessions, and expose lifecycleState on SessionSummary for future sidebar rules. Completes the #833 P0 follow-up alongside agent-id dedup. Fixes #833 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): scope sidebar dedup key by flavor Prevent cross-flavor collisions when flattened agentSessionId retains a stale native id. Add regression test and relax claudeRemote CI timeout. Fixes #833 Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
ad038bbf2e |
fix(cursor): merge SKU catalog under ACP lock and refcount agent guard (#835)
Fixes incomplete cliModelSkus while agent acp holds the CLI lock (#831) and replace single-pid ACP lock with cross-process refcount (#832). Web picker merges machine/session catalogs and waits for SKU readiness before showing variant labels. Fixes #831 Fixes #832 Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
HeavyGeeandGitHub
|
6d2d0d4707 |
fix(cursor): trim #784 safety patch to marker-only on legacy stream-json path (closes #822) (#828)
* fix(cursor): drop timing heuristic from #784 intercept; scan raw payload (#801 follow-up) PR #801 shipped a two-strategy intercept for the synthetic AskQuestion skip response in legacy stream-json mode. Real-traffic data from a post-merge run shows the marker-match strategy never fires (the converter's `extractToolResult` discards the marker for tool shapes it does not recognize, returning `{}`) and the timing-signature defense-in-depth strategy fires only on false positives - notably the Anthropic Vertex Claude tool calls cursor-agent surfaces in legacy sessions, which all land as `name=unknown` with the `{}` extracted result and frequently complete under the 500 ms threshold. Measured on a single legacy-resumed session (`7b769423`): 1,136 `name=unknown` tool calls, 16 rewritten as `no_input_surface`, zero actual marker strings stored anywhere in the session. The 16 rewrites were legitimate fast tool calls (Anthropic Vertex `toolu_vrtx_*` IDs) mischaracterized as fabricated skip responses. Changes: - Remove the timing-signature heuristic and its supporting state (started-at map, elapsed-ms calculation, latency threshold, test-only state reset). - Move the marker scan from the post-`extractToolResult` output to the raw `tool_call` payload, so it can see the marker on stream-json shapes the converter does not specifically recognize. Function-shaped tools exclude `function.arguments` from the scan to avoid matching agent-controlled input. Other shapes scan the full payload (no agent-input field exists at the top level). - Refresh tests: drop timing-based positive cases, add a marker-in-raw- payload positive case for `name=unknown` shapes, and add a regression that legitimate fast `name=unknown` tool calls without the marker pass through with `status: completed`. - Document scope: this intercept now lives only on the legacy stream- json path, which only resumed pre-ACP sessions hit. New cursor remote sessions go through `cursorAcpBackend` and the `cursor/ask_question` ACP extension method (#799) - immune to this bug. The intercept drains with the legacy session population. Tracking: #784. Builds on #801, complements #799. * fix(cursor): exclude agent input from marker scan; surface top-level Anthropic tool names (Codex P2) Codex flagged a false-positive case on the fork-stage review of this branch (heavygee/hapi#35, P2): an Anthropic tool_use shape with a top-level `name` (e.g. `{id, name: 'TodoWrite', input: { ... }}`) gets labelled `name=unknown` by the converter and passes the AskQuestion gate. If the agent's `input` quotes the synthetic-skip marker - which happens whenever an agent edits or documents this very bug - the intercept would rewrite a perfectly fine TodoWrite as a fabricated skip. Two-part fix: 1. `extractToolName` now reads the top-level `name` field as a final fallback. A real `TodoWrite` / `Bash` / `str_replace_based_edit_tool` surfaces with its actual name and is rejected by the AskQuestion gate before the marker scan runs. The original AskQuestion fabrication case still surfaces as `unknown` (per #784 issue body the name is stripped in the fabricated payload) and remains detectable. 2. Defense in depth: introduce `AGENT_INPUT_KEYS = {input, args, arguments}` and exclude these from the non-function shape's marker scan. Even if a tool reaches this code path with `name=unknown` and the marker buried in its `input`, the intercept won't fire on agent- controlled text. Two new regression tests: - Anthropic tool_use shape `{id, name: 'TodoWrite', input: {todos: [ marker]}}` → passes through with `status: 'completed'`. - `name=unknown` shape with marker only inside `input` → passes through with `status: 'completed'`. All 20/20 tests pass; typecheck clean (cli + web + hub). |
||
|
|
cb72703649 |
feat(hub+web): add POST /sessions/:id/reopen + Reopen button on inactive rows (#826)
* feat(hub+web): add POST /sessions/:id/reopen + Reopen button on inactive rows
Archived sessions retain their full transcript and metadata in the DB, but
today there is no path back to them from the web UI; the only way to revive
one is shell access plus sqlite metadata patching plus a manual /resume call.
This change adds a single one-click affordance:
- Hub: new POST /api/sessions/:id/reopen route on the existing sessions
router. The route delegates to a new engine method `reopenSession` that:
- is idempotent (active session -> 200 with `resumed:false`),
- validates Cursor sessions still have a `cursorSessionId` once they have
any messages (otherwise we cannot resume the agent thread),
- clears `lifecycleState='archived'`, `archivedBy`, `archiveReason` via a
versioned metadata update, and stamps `lifecycleStateSince`,
- defaults `cursorSessionProtocol='stream-json'` for pre-#799 Cursor
sessions (sessions that have a `cursorSessionId` but no protocol set),
so routing still reaches the legacy launcher; ACP sessions keep their
explicit protocol,
- forwards to the same `resumeSession` path the existing /resume route
uses, including the `canFreshSpawnNeverStartedSession` fallback.
422 is returned with `{ missing: [...] }` when the agent metadata needed
to resume is gone; other engine errors map to 404/409/503/500 with the
existing shape (mirrors /resume).
- Web: a "Reopen" entry in the SessionActionMenu that appears next to
"Delete" on inactive sessions only. Wired into both the SessionList rows
and the SessionHeader more-menu, with a small dismissable error dialog
for the 422 missing-metadata case.
- Tests: route-level coverage for the four response shapes (200 reopen,
200 idempotent, 404, 422) plus 409/503 error mappings; sessionCache
tests for the archive-metadata clear (including the legacy Cursor
protocol default); React component test for the menu item rendering on
inactive vs active sessions; mutation hook test for the api wiring and
the ApiError surface needed by the UI.
Closes #819
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(reopen): address codex review findings on fork PR #33
Four P2 findings from the cold-review bot, three fixed and one explained:
1. Mutation now returns the reopen response so the UI can route to a possibly
different sessionId. SyncEngine.resumeSession may merge the row into a
freshly-spawned session id (matching the send-message resume flow); the
chat view now navigates there, the row list calls onSelect on the new id.
2. reopenSession on the client now goes through `request()` instead of a
hand-rolled fetch, so 401 + onUnauthorized refresh works the same as
every other session action. `request()` now throws `ApiError` (with
status/code/body) on non-401 errors - backward compatible because
ApiError extends Error.
3. (Reply only) Pre-#799 Cursor protocol propagates correctly without the
extra plumbing the bot suggested: `clearSessionArchiveMetadata` writes
`cursorSessionProtocol='stream-json'` to the DB; the CLI's
`bootstrapExistingSession` preserves it via `pickExistingSessionMetadata`;
if it's still absent at the launcher, `isLegacyCursorSession` defaults
to stream-json whenever `cursorSessionId` is present.
4. Archive metadata is now restored when resume fails. `reopenSession`
captures a snapshot of `lifecycleState`/`archivedBy`/`archiveReason`/
`lifecycleStateSince` before the clear; if `resumeSession` returns an
error (no machine online, spawn timeout, etc.), the snapshot is put
back via the new `SessionCache.restoreSessionArchiveMetadata`. Engine
test covers both the rollback and the no-rollback-on-success cases.
Error rendering helper moved to `web/src/lib/reopenError.ts` so the chat
header and the session row share one implementation, and gained a unit test.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(web): preserve engine error codes in ApiError.code on /reopen
`/sessions/:id/reopen` returns `{ error, code }` where `code` is the stable
taxonomy (`no_machine_online`, `resume_unavailable`, etc.) and `error` is the
human-readable message. The generic `request()` error path was reading only
`parsed.error`, so `ApiError.code` ended up being a message like
"No machine online" rather than `no_machine_online`, breaking taxonomy-based
branching in web callers.
`parseErrorCode` now prefers `parsed.code` and falls back to `parsed.error`
for legacy routes that only set `error`. Added api/client.test.ts covering
the three response shapes /reopen actually emits (503 with code, 500 without
code, 422 with missing[]).
Addresses upstream codex-action review on tiann/hapi#826.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(reopen): restore archive metadata exactly on rollback (drop fresh lifecycleStateSince)
For an archived session that predates `lifecycleStateSince` (the field is
absent from its metadata), `clearSessionArchiveMetadata` stamps a fresh
timestamp. If `resumeSession` then fails, the rollback was leaving that
fresh timestamp in place, making the rolled-back row look like it was
just archived rather than preserving the original lifecycle age.
`restoreSessionArchiveMetadata` now does an EXACT restore: when a snapshot
field is undefined the corresponding key on the metadata is deleted, not
left alone. Applies symmetrically to lifecycleState / archivedBy /
archiveReason / lifecycleStateSince. Test updated to assert the deletion
of the fresh timestamp.
Addresses upstream codex-action review on tiann/hapi#826.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
|
||
|
|
cd99cfbc25 |
fix(hub): preserve session metadata across archive transitions (#825)
* fix(hub): preserve flavor session ids in metadata across archive transitions When a session ends (terminate, crash, local-launch failure, handoff), the runner's archive write replaces sessions.metadata wholesale. If the CLI's locally cached Metadata is null (e.g. Zod parse failed at bootstrap and api.ts nulled it out) or stale, the spread in archiveAndClose ships a sparse blob and the resume token (cursorSessionId, codexSessionId, claudeSessionId, etc.) gets cleared from the row even though the on-disk chat data is still intact. Fix at the hub layer because update-metadata is the single chokepoint for every metadata write surface (CLI, web, future): in the store-level updateSessionMetadata, read the prior row's metadata inside a transaction and carry forward a small allowlist of flavor resume tokens when the incoming write omits them. Explicit overwrites still win. The allowlist mirrors pickExistingSessionMetadata in sessionFactory.ts which already preserves the same fields on bootstrap. Closes tiann/hapi#820 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(hub): address cold-review findings on metadata merge Three bot findings on the initial patch: 1. (P1) Sparse archive payloads still resulted in metadata blobs that failed MetadataSchema parse downstream — required `path`/`host` were not in the carry-forward set, so even though the resume token survived, hub session cache and CLI getSession nulled-out the row and resume_unavailable came back. Add PARSE_IDENTITY_FIELDS = `path`, `host` to the carry-forward. 2. (P2) Preserving `cursorSessionProtocol` whenever it was omitted carried a stale protocol over to a freshly written `cursorSessionId`, misrouting a future remote resume. Pair-aware logic: drop the prior protocol when next sets a new id; preserve the protocol only when next is silent on both id and protocol. 3. (P2) The successful update-metadata broadcast emitted the pre-merge payload to other CLIs in the session room, so even though the DB row was preserved, peer caches diverged. Switch the broadcast value to `result.value` (the persisted merged value) so live caches stay in sync with the truth. Refactor preserveProtocolResumeFields into mergeSessionMetadata with two tiers (PARSE_IDENTITY_FIELDS + SIMPLE_RESUME_TOKENS) plus the cursor pair handler. 6 new tests cover the regressions; existing 16 still pass plus 1 new socket-level test for the broadcast. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(hub): preserve flavor + machineId across sparse metadata merges Bot P2 on the prior fix: PARSE_IDENTITY_FIELDS (path, host) made the blob parseable and SIMPLE_RESUME_TOKENS preserved the chat-id, but flavor and machineId were still being dropped by sparse archive payloads. Consequences: - flavor: hub/src/web/routes/sessions.ts and sync/syncEngine.ts read `metadata?.flavor ?? 'claude'` to pick which session id field to resume. With flavor missing, a Cursor/Codex/Gemini session was routed as Claude and the preserved cursorSessionId was ignored. - machineId: telegram/bot.ts and the CLI's resumable listing read `metadata?.machineId` to scope sessions to the current host. With machineId missing, the row dropped out of the resume picker. Add a third carry-forward tier ROUTING_FIELDS = [flavor, machineId] between PARSE_IDENTITY_FIELDS and SIMPLE_RESUME_TOKENS in mergeSessionMetadata. 3 new tests cover preservation, no-invention, and explicit override. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(hub,cli): support explicit-clear sentinel for carry-forward fields Upstream cold-review (Major): the carry-forward semantics introduced in the prior commits ("omit field → preserve from prior") collide with cli/src/codex/session.ts resetCodexThread(), which intentionally clears codexSessionId by deleting it from the metadata blob before calling updateMetadata. With omit-as-preserve, the cleared id was restored from the prior row and /clear on a Codex session no longer dropped the persisted thread. Add an explicit-clear sentinel: when next sets a carry-forward field to `null`, the merge drops the key entirely from the persisted blob (key removed; not stored as null since MetadataSchema fields are `string().optional()`). `undefined` (key missing from next) keeps its "carry forward" meaning. The two semantics now compose cleanly: - next.field = "x" → next wins (caller sets a new value) - next.field = null → drop the field (caller intentionally clears) - next omits field → carry forward prior (caller didn't touch it) Update resetCodexThread() to send `codexSessionId: null` so the reset actually drops the persisted thread under the new merge. 4 new hub tests cover: explicit clear of a single token, clear-one- preserve-others independence, no-op clear on a never-set field, and the success-ack value reflects the cleared blob. cli/src/codex tests (224/224) and hub suite (301/301) green; bun typecheck clean. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
edc3acc3e2 |
fix(cursor): requeue user message on transient agent exit (auth, rate limit) (#823)
* fix(cursor): requeue user message on transient agent exit (auth, rate limit)
cursorLegacyRemoteLauncher.runMainLoop popped a user message off the queue
before spawning `agent` and silently discarded it whenever `agent` exited
non-zero (auth expiry, rate limit, transient network). The wrapper logged
the failure at debug level only, never surfaced it to the web UI, and
emitted `ready` as if a normal turn had ended.
Capture stderr from the spawned process; classify exit-1 with a transient
signature (Authentication required, rate limit, ETIMEDOUT, ECONNRESET,
EAI_AGAIN) as recoverable; re-head the message via `queue.unshift`, surface
a friendly banner via `sendSessionEvent({type:'message',...})`, and backoff
~2s before the loop picks it up again. Cap at 5 consecutive transient
failures, after which the message is dropped with a clear "resolve and
resend" event so we never spin forever on a genuinely broken auth.
Non-transient non-zero exits also surface the stderr to the UI now (instead
of only the local ring buffer), so a real crash is visible to the operator.
Backoff is overridable via CURSOR_LEGACY_TRANSIENT_BACKOFF_MS for tests.
Tests cover: success path, transient auth requeue + banner, rate-limit
banner, non-transient crash surfaced without requeue, and the 5-failure
drop cap.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(cursor): preserve slash-command isolation on requeue; wait for stderr flush
Two findings from the cold-review bot on the requeue path:
1. `enqueueCursorUserMessage` uses `pushIsolated` for pass-through slash
commands (e.g. `/compress`) so they never batch with sibling prompts.
The transient-requeue path used plain `unshift`, which dropped the
isolate bit and allowed the next collected batch to merge the slash
command with a sibling - changing command semantics. Add
`MessageQueue2.unshiftIsolated` and use it when the popped batch was
isolated or when `parseCursorSpecialCommand` recognises the message.
2. `runAgentProcess` resolved on `child.on('exit', ...)`. Node may emit
`exit` while the stderr pipe is still draining, so a fast "auth
required" error printed-and-exited could be classified as
non-transient with empty stderr and silently drop the user message -
the exact bug this PR was supposed to fix. Resolve on `close` instead,
which waits for stdio streams to flush.
Adds a unit test that requeues `/compress` after a transient auth failure
and asserts the second spawn still receives the slash command alone (not
batched with a sibling).
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(cursor): restrict transient retry to exit code 1 only
Upstream codex review #823 (Minor): the helper treated any non-zero exit
with matching stderr as transient, which could requeue a signal-killed
(SIGTERM 143, SIGKILL 137) or crashed (SIGABRT 134) process whose stderr
happens to contain a keyword like "rate limit". Documented contract is
exit-1-for-transient; tighten the classifier accordingly.
Adds regression test covering exit 143 + rate-limit stderr → no retry.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(cursor): clean up transientBackoff abort listener on timer completion
Upstream codex review #823 (Minor): transientBackoff added an abort
listener with { once: true } but only removed it when the abort fired.
Because the launcher reuses one AbortController, repeated transient
retries accumulated stale listeners until the next abort.
Switch to a single completion path that clears the timer AND removes the
abort listener whichever side wins.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(cursor): cap in-memory stderr capture at 8 KB
Upstream codex review #823 (Minor): runAgentProcess accumulated every
stderr chunk for the full child lifetime. A noisy `agent` failure could
grow CLI process memory without bound even though only the first 400
chars are ever displayed. Cap the retained copy at 8 KB; debug log of the
full stream is unchanged.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
|
||
|
|
9349acba62 |
fix(web): use session flavor label in voice context formatters (closes #680) (#815)
Replaces hardcoded \"Claude Code\" strings in voice context injections with the active session's flavor label (Cursor, Codex, Gemini, etc.) via getFlavorLabel() from @hapi/protocol. Falls back to \"coding agent\" for unknown or missing flavors. Threads an agentLabel string param through: - formatMessage - formatNewSingleMessage - formatNewMessages - formatHistory - formatSessionFull - formatPermissionRequest - buildSessionVoiceContextPlan (passes to its internal formatMessage call) voiceHooks adds a single getAgentLabel(session) helper that reads session.metadata.flavor and resolves the display label once per call. Tests updated to cover the new parameter shape and to assert that \"Claude Code\" is never substituted regardless of agent flavor. formatReadyEvent already used a generic \"coding agent\" phrasing and needs no signature change. Closes #680 Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
c87f88f02f |
fix(web): disable single-dollar inline math in remark-math (#805)
* fix(web/markdown): disable single-dollar inline math in remark-math Default remark-math configuration treats $...$ as inline LaTeX, which turns any prose containing two currency amounts (e.g. "save $400 vs the $200 plan") into a KaTeX block — paragraphs collapse, whitespace is stripped, the running text is re-rendered as math symbols. Pass `singleDollarTextMath: false` to remarkMath so single $ is plain text. Block math `$$...$$` (on its own line) still renders, matching GitHub-flavored markdown semantics. Single source of truth: MARKDOWN_PLUGINS is shared by MarkdownText, Reasoning, and MarkdownRenderer — fix lands in all three surfaces. Adds 3 regression tests that drive the unified pipeline end-to-end: prose with multiple "$N" amounts produces no `class="katex"` and no `<math>` element; `$$...$$` block math still does. Co-authored-by: Cursor <cursoragent@cursor.com> * chore(web): declare unified/remark-parse/remark-rehype/hast-util-to-html The new markdown-text regression test imports these directly to drive the unified pipeline end-to-end. They were resolving via transitive deps from remark-math and rehype-katex, which is fragile — a future dep upgrade can remove the transitives and break the test. Declare them explicitly under devDependencies. No code change; lockfile records the same versions that were already installed transitively (unified@11.0.5, remark-parse@11.0.0, remark-rehype@11.1.2, hast-util-to-html@9.0.5). Addresses the HAPI Bot review finding on PR #805. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
SSU-WEI HUANGandGitHub
|
66ba312117 |
fix(web): suppress Mermaid render error SVGs (#813)
* test: reproduce issue #785 * fix: suppress Mermaid render errors (closes #785) |
||
|
|
3a8693f380 |
feat(cursor): migrate remote sessions to ACP with model/variant pickers (#799)
* feat(cli,web,hub): migrate Cursor remote sessions to ACP with model/effort pickers Move stream-json remote launcher to legacy path and add ACP launcher with set_config_option model/mode sync, optimistic keepalive on config changes, and shared catalog caching. Web gets dual base/effort Cursor pickers for session and new-session flows; hide composer status bar when Cursor sends no usage_update. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(cli,web,shared): Cursor model picker — ACP wires + CLI sku variants Enrich the web/mobile picker with agent --list-models SKUs grouped under ACP wire bases, fix session-open base highlight, and keep catalog discovery safe while the ACP transport holds the CLI lock. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(cursor-acp): apply ACP default model when web resets to Default Web sends model: null for Default; push session/set_config_option with the ACP default[] wire so Cursor backend matches hub state. Regression tests for setModel(null) and applyModelConfig(null). Co-authored-by: Cursor <cursoragent@cursor.com> * fix(acp): clear stale agent-acp lock when owning process is gone Check lock pid with signal 0; remove orphaned lock dirs after SIGKILL or crash so listCursorModels can run cold probes again. Regression tests for guard and catalog discovery. Co-authored-by: Cursor <cursoragent@cursor.com> * test(cursor): use live pid for ACP lock handler tests Stale-lock cleanup clears dead pids; handler tests must simulate an active lock with the current process pid to avoid cold probes/timeouts. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(acp): scope agent CLI lock guard to Cursor agent command only Gemini/OpenCode/Kimi ACP sessions must not register agent-acp-active; that blocked listCursorModels while unrelated backends were running. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(hub,web): reject Cursor model changes for local sessions Hub returns 409 when controlledByUser is set, matching Codex. Web hides model and variant pickers for local Cursor sessions so users do not hit a dead RPC path. Document pre-push-review in AGENTS.md. Verified: bun typecheck; bun run test (919 cli + 243 hub + 768 web + 46 shared). Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): send stable ids for Cursor ask_question replies Parse and submit question.id and option.id so ACP receives keys like { approach: ['a'] } instead of index/label. Verified: bun typecheck && bun run test. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
hangerandGitHub
|
f4c3513abd | fix(web): drop stale queued-message ghosts on at-bottom refresh (#811) | ||
|
weishu
|
d82ff6127d | Release version 0.20.0 | ||
|
HeavyGeeandGitHub
|
dc0d21e05b |
fix(cursor): intercept fabricated Questions skipped AskQuestion result in headless mode (#784) (#801)
* fix(cursor): intercept fabricated 'Questions skipped' AskQuestion result in headless mode (#784) When cursor-agent runs under `--print --output-format stream-json` (HAPI's current Cursor remote launcher), the CLI returns a synthetic `Questions skipped by the user, continue with the information you already have` response for the `AskQuestion` tool in ~zero seconds with no error flag, because there is no IDE surface to render the question. The underlying model can interpret this as legitimate user consent and act on it. This patch intercepts the synthetic result in `cli/src/cursor/utils/cursorEventConverter.ts` and rewrites the `tool_call`/completed event to a structured `no_input_surface` failure (`status: 'failed'`, which downstream becomes `is_error: true`). Detection has two strategies: 1. String match - any `tool_call`/completed payload whose serialized form contains the synthetic-skip marker is rewritten. This is robust to wherever cursor-agent stuffs the marker inside the `tool_call` object. 2. Timing + name heuristic (defense in depth) - any completion that arrives within 500 ms of its 'started' event with a trivial result, for a tool call named `AskQuestion`, `askQuestion`, `ask_question`, or the converter's `unknown` fallback, is also rewritten. This catches the case where cursor-agent changes the synthetic-string text in a future release. The converter tracks per-call timestamps in a bounded `Map` (`<= 1024` entries, oldest evicted on overflow) and clears entries when the corresponding 'completed' event arrives. A small test-only reset hook isolates state between Vitest cases. This is a transitional safety patch. It auto-deletes when #781's ACP launcher replaces the stream-json launcher and `cursor/ask_question` becomes a proper bidirectional ACP method where fabrication is structurally impossible. Scope is intentionally tiny: only `cli/src/cursor/utils/cursorEventConverter.ts`, its colocated Vitest file, and a section in `docs/guide/cursor.md`. No changes to `cursorRemoteLauncher.ts`, ACP code, web normalizer, or permission UI. Refs: tiann/hapi#781 (long-term resolution via ACP migration) Closes: tiann/hapi#784 * fix(cursor): gate AskQuestion intercept on tool name (#784 PR #801 review) Address regression flagged by the HAPI auto-review bot on #801: `containsSyntheticSkipMarker` previously stringified the entire `tool_call` payload and matched the literal marker substring. Because this PR also adds that exact marker to `docs/guide/cursor.md` (to document the intercept), a Cursor `read_file` of that documentation page would surface the marker inside `readToolCall.result.content` and be rewritten as a `no_input_surface` failure, corrupting an unrelated, legitimate result. The intercept is now gated on the tool name resolving to an AskQuestion-shaped call (`AskQuestion`, `askQuestion`, `ask_question`, or the converter's `unknown` fallback for unnamed function-shaped tools). `read_file` / `write_file` tool calls - which have explicit `read_file` and `write_file` names from `extractToolName` - no longer fall under the intercept, regardless of what their payload contains. The marker check itself now walks values recursively (string / array / object), guarded by a `WeakSet` against cycles, instead of relying on `JSON.stringify`. Slightly tidier; behaviour is otherwise unchanged for the AskQuestion path. Regression tests added: - `read_file` result whose `content` contains the marker -> passes through with `status: 'completed'` and no `no_input_surface`. - `write_file` whose serialized `args` contain the marker -> same. - A non-AskQuestion function tool (`MyCustomTool`) whose result quotes the marker -> same. All 846 cli tests pass (17 in this file). `bun run typecheck` exits 0. * fix(cursor): scope synthetic-skip check to extracted result (#784 PR #801 review-2) Address second Major finding from the HAPI auto-review bot on #801: After the previous fix gated the intercept on the tool name, the marker check still recursed into the entire `tool_call` object - which includes `function.arguments`, the agent's own prompt text. A legitimate AskQuestion whose prompt quotes the synthetic-skip marker (e.g. an agent debugging this exact bug, or any prompt that pastes the marker verbatim) would have been rewritten as `no_input_surface` even when the operator actually answered. Changes: 1. `extractToolResult` now extracts the cursor-side response from function-shaped tool calls. Previously it returned `{}` for anything that wasn't `readToolCall` or `writeToolCall`. It now returns `function.result` when present, otherwise every field of `function` except `name` and `arguments`. This excludes the agent's input from what downstream sees as the tool result, and as a side effect surfaces the actual cursor response for function-shaped tools (which was previously lost - see the #784 incident note about HAPI storing `output: {}` for AskQuestion in the message DB). 2. `shouldRewriteAsNoInputSurface` now searches only the extracted `result`, not the whole `tool_call`. The bot's exact recommendation. 3. Test added: an AskQuestion whose `arguments` quote the marker but whose `result` is a real user answer, with elapsed time past the 500 ms threshold so the timing heuristic does not apply. Asserts the tool_result passes through with `status: 'completed'` and the operator's actual answer. All 847 cli tests pass (18 in `cursorEventConverter.test.ts`). `bun run typecheck` exits 0. The widened `extractToolResult` scope is necessary for the marker check to actually find the synthetic string (it lives inside `function.result` or a sibling field), and is the bot's explicit recommendation. It also removes the long-standing data-loss bug where AskQuestion responses were surfaced to the message DB as opaque `{}` - regardless of fabrication. |
||
|
|
a812a51dd7 |
feat(voice): backend voice picker + advanced controls behind disclosure (#742) (#743)
* feat(voice): voice personality, picker catalog, and prompt layer foundation - voicePickerCatalog.ts: per-backend voice lists for Gemini and Qwen with resolve helpers (resolveGeminiLiveVoice, resolveQwenRealtimeVoice) - voicePersonality.ts: VoicePersonalityPreferences schema, presets, composed system prompt with identity/character/response-length layers - voicePromptLayers.ts: buildResolvedVoiceSystemPrompt, preset delivery snippets - voiceSystemPromptParam.ts: hub-side base64url decode for ?systemPrompt= - voicePickerPreferences.ts, voicePersonalitySession.ts: browser-side encode, decode, and storage helpers - useVoicePersonality: React hook for preferences persistence via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): preset delivery included when non-balanced preset selected; restore test typecheck - isDefaultVoicePersonality: add preset check so warm/calm/direct presets trigger the delivery snippet instead of being treated as default - web/tsconfig.json: remove test file exclusion from typecheck (restoring strict coverage of test code); fix resulting type error in mock declaration via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): include use_speaker_boost in ElevenLabs TTS override payload The checkbox persisted the pref but ttsDiffersFromDefault and buildElevenLabsTtsOverride both omitted it, so the setting was never sent to the agent. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * test(voice): update speaker_boost test to assert it IS included in override The previous test asserted use_speaker_boost was omitted; now it's correctly included in the TTS payload. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): authorize use_speaker_boost in ElevenLabs override schema Add use_speaker_boost to both the VoiceAgentConfig tts override type and the buildVoiceAgentConfig() platform_settings so the field is accepted by the ElevenLabs agent runtime. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): propagate full language code through composed prompt, not just zh getDefaultVoiceSystemPrompt and resolveComposedVoiceSystemPrompt were filtering language to zh-only before passing to composeVoiceAgentPrompt. Now append buildVoiceLanguageBlock(language) after composition so French, Spanish, Japanese etc. reach Gemini/Qwen sessions correctly. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): only append language block when language explicitly set Building language block unconditionally when no language is given caused getDefaultVoiceSystemPrompt() to diverge from VOICE_SYSTEM_PROMPT. Only append the block when a code is explicitly provided. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): always include language block for Gemini/Qwen in composed prompt When auto-detect is on (language=undefined), the composed prompt sent via hub proxy was losing the language auto-detect instruction because the block was only added when language was explicitly set. Now: ElevenLabs skips the block (has its own language field); Gemini/Qwen always include it — undefined produces the auto-detect block, an explicit code produces the appropriate language instruction. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> --------- Co-authored-by: HAPI <noreply@hapi.run> |
||
|
SSU-WEI HUANGandGitHub
|
d09168778c |
fix(web): preserve user prompt line breaks (#804)
* test: reproduce issue #794 * fix: preserve user prompt line breaks (closes #794) |
||
|
|
f086949a8a |
feat(web,hub): export session conversation (#808)
* test: reproduce issue #793 * fix: add session conversation export (closes #793) * fix(hub): sort session export by display time for invoked scheduled messages Export now uses COALESCE(invoked_at, created_at) ordering so JSON/Markdown exports match the visible chat chronology after scheduled messages are invoked. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): escape newlines in session export YAML front matter Prevent session metadata containing newlines or quotes from breaking Markdown export front matter. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
dbc9646d06 |
Show and filter work directories in Codex import (#810)
* Add workdir filter to Codex import dialog * Fix filtered select all in Codex import --------- Co-authored-by: userZ <spare.nets-0y@icloud.com> |
||
|
|
59c29e8423 |
feat(voice): pluggable voice backend with Gemini Live & Qwen Realtime (#692)
* feat(voice): pluggable voice backend with Gemini Live & Qwen Realtime Rebased from Overbaker/hapi#401 onto current main. Adds a pluggable voice backend architecture that extends the existing ElevenLabs integration: - Gemini 2.5 Live (gemini-live): Google real-time audio via WebSocket with full function calling (messageCodingAgent, processPermissionRequest) - Qwen Realtime (qwen-realtime): Alibaba DashScope via hub WebSocket proxy (browser cannot set Authorization header directly) - VoiceBackendSession: dynamic backend selector with React.lazy loading, gates voice button until backend module is registered - Hub WS proxies: JWT-authenticated /api/voice/gemini-ws and /api/voice/qwen-ws endpoints in Bun.serve, with message queueing during upstream connect to prevent dropped setup frames - AudioWorklet pipeline: inline Blob URL recorder, 24 kHz PCM player, serial tool call execution, AudioContext created in user gesture for mobile - Backend discovery: GET /voice/backend + POST /voice/gemini-token / POST /voice/qwen-token hub routes; frontend auto-detects active backend Merge notes: - Rebased 135 upstream commits cleanly; HappyComposer keeps upstream's configurable enter-behavior setting (supersedes hard-coded Ctrl+Enter) - Converted gemini test files from bun:test to vitest (web package uses vitest) - All 221 hub tests and 636 web tests pass; TypeScript clean * fix(voice): restore user mic mute state after Gemini turn completes turnComplete handler was unconditionally calling setMuted(false), which re-enabled the mic track even when the user had manually muted. Now restores to state.micMuted instead. * fix(voice): remove hard-coded Chinese language from Gemini backend buildGeminiLiveConfig was appending VOICE_CHINESE_LANGUAGE_BLOCK which forced Gemini to always respond in Mandarin regardless of user locale. Gemini now uses the neutral base prompt and responds in the language the user speaks to it, consistent with the ElevenLabs behaviour. * fix(voice): reset modelSpeaking in cleanup to unblock mic on restart If the session closes while Gemini is mid-speech, cleanup() left state.modelSpeaking=true. The next startSession() would then drop all mic audio in sendAudioChunk() until a model turn eventually flipped the flag — effectively deaf until page reload. * fix(voice): guard stale close handlers in Gemini and Qwen sessions ws.onclose operated on module-level state.ws, not the socket that fired the event. A rapid stop/restart could cause the old socket's onclose to call cleanup() after the new socket was assigned, tearing down the live session. Guard with `if (state.ws !== ws) return` before cleanup. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): remove hard-coded Chinese language from Qwen backend Matches the Gemini fix — both backends now use VOICE_SYSTEM_PROMPT without the Chinese language block, giving consistent English-default behaviour across all non-ElevenLabs backends. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * feat(voice): proactive/reactive toggle in voice settings Adds a "Proactive voice" toggle (default: off = reactive) to the Voice Assistant settings section. Reactive (default): initial context and agent-ready events are fed silently; the assistant waits for the user to speak first. Proactive: original behaviour — Gemini/Qwen narrate context on connect and speak unprompted when the agent finishes a task. ElevenLabs is also affected via onReady sending a user message rather than a silent update. Covers all three backends uniformly. localStorage key: hapi-voice-proactive. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): normalize WS close codes, drop barrel re-exports, fix SSE visibility - hub/server.ts: add toClientCloseCode() to normalize reserved upstream close codes (1005/1006/1015) to 1011 before forwarding to browser; abnormal upstream drops (1006) would otherwise throw on clientWs.close() and leave the browser socket open - realtime/index.ts: remove static GeminiLiveVoiceSession and QwenVoiceSession barrel exports; VoiceBackendSession lazy-imports both, so barrel re-exports created static dependencies that defeated the intended code-split - App.tsx: gate global useVisibilityReporter on !sessionEventSubscription so the always-on SSE connection does not suppress native Web Push notifications for sessions the user is not currently viewing via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): respect language setting in Gemini/Qwen; fix voice-start toggle label - buildGeminiLiveConfig() now accepts optional language param; appends VOICE_CHINESE_LANGUAGE_BLOCK only when language === 'zh' - GeminiLiveVoiceSession passes config.language through - QwenVoiceSession conditionally builds basePrompt from language setting - Fixes silent no-op when user selects Chinese in voice settings on Gemini/Qwen backends (was ElevenLabs-only) - Rename voice-start toggle label to 'Start voice session with summary' - Fix description: clarifies the choice is about session-open behaviour (summary vs greeting), not ongoing narration via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): send greeting trigger in reactive mode for Gemini Gemini Live has no built-in first-message like ElevenLabs agents do; without an explicit turnComplete:true it sits silently. In reactive mode (default, toggle off) now sends a greeting instruction after any silent context feed so Gemini introduces itself and invites the user to speak. Proactive mode is unchanged: the context summary is the opening speech. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): suppress Gemini self-identification and context leak in greeting - VOICE_SYSTEM_PROMPT: explicit instruction never to call itself Gemini, Google, or any underlying model/provider name — always HAPI - Greeting trigger text: instruct to greet as HAPI only, suppress model name and any reference to context/recent activity in the opening line via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): address code review findings — error handling, proxy, audio Gemini + Qwen client: - onerror now sets setupDone/sessionReady and nulls state.ws before calling reject(), so the stale-close guard trips in onclose and prevents a duplicate statusCallback('error') on WS failure Gemini client: - Proactive mode with no initialContext now falls through to the greeting trigger instead of sitting silently - Remove unused handleBargeIn callback (dead code) Qwen client: - Add input_audio_sample_rate: 16000 to session.update so PCM rate is declared explicitly rather than relying on DashScope's default Hub proxy: - Remove no-op ternary in Gemini flush loop and message handler (typeof x === 'string' ? x : x); use upstream.send(msg) directly - Qwen onerror now calls upstreamMap.delete() before closing client, eliminating the stale map entry window - Align Qwen hub fallback model string with QWEN_REALTIME_MODEL constant ('qwen3-omni-flash-realtime') via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): trailing-slash WS URL, Qwen session.update schema hub/voice.ts: - Replace string-concat WS URL construction with buildVoiceWsUrl() which uses URL API to set protocol/pathname cleanly — fixes double-slash when HAPI_PUBLIC_URL has a trailing slash (would silently skip the proxy route) QwenVoiceSession.tsx: - Wrap tool definitions in {type:'function', function:{...}} as required by Qwen-Omni realtime schema — previous flat shape caused session.update rejection before audio capture could start - Use pcm16/pcm24 audio formats matching DashScope spec; remove input_audio_sample_rate (encoded in format name) via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): await audio capture before setMuted; sanitize upstream close codes GeminiLiveVoiceSession + QwenVoiceSession: - startAudioCapture() is now async and awaits recorder.start() before calling setMuted() — previously setMuted ran before getUserMedia resolved so a session restarted while muted would open the mic anyway - statusCallback('connected') now fires after audio is ready - setMuted() called unconditionally (not just when true) to correctly apply saved state in either direction hub/src/web/server.ts: - Both Gemini and Qwen close() handlers now pass the client code through toClientCloseCode() before forwarding to upstream — prevents reserved codes (e.g. 1006) from causing WebSocket.close() to throw and leave the upstream session open until provider timeout - Reason string capped at 123 bytes (WebSocket protocol limit) via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): wrap startAudioCapture in try/catch to propagate mic errors An unhandled rejection inside the async onmessage callback does not propagate to the outer startSession Promise — the UI hangs on 'connecting' and the provider socket stays partially open. Wrapping the await in try/catch calls cleanup()/statusCallback('error')/reject() so failures surface correctly. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): propagate backend discovery failure instead of silently falling back to ElevenLabs fetchVoiceBackend no longer catches errors and defaults to 'elevenlabs' — any network or server failure now throws so VoiceBackendSession can surface it via onStatusChange('error', ...) rather than silently mounting the wrong backend. VoiceBackendSession also resets backend state to null when api changes, so a stale ElevenLabs registration from a prior discovery cannot persist into a new session. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): throw on unrecognised backend value instead of silently falling back to ElevenLabs Unknown backend strings (future values, typos) now throw rather than defaulting to elevenlabs, closing the narrow remaining form of the original misrouting bug. Also removes the unnecessary `as VoiceBackendResponse` cast. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): add Qwen greeting/proactive trigger; fix socket buffer for base64 uploads Qwen session.updated handler now sends the same proactive summary or greeting trigger that Gemini does — previously it started silently in both proactive and reactive modes. maxHttpBufferSize raised to 68 MiB to account for base64 expansion: 50 MiB decoded files become ~66.7 MiB as base64 JSON, so the previous 55 MiB ceiling would disconnect uploads above ~41 MiB before they reached the CLI. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): replace unsupported conversation.item.create with session.update for Qwen text Qwen's realtime API only supports conversation.item.create for function_call_output. Sending it with type:'message' for greetings/context was invalid and could fail before the user spoke. sendTextMessage and sendContextualUpdate now update session instructions via session.update (accumulating context into the system prompt) and trigger response.create only when a spoken reply is needed — matching Qwen's supported client event surface. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): guard session.updated re-entry and reset config on session start session.updated now returns early after the first ack — subsequent session.update calls (instruction appends) also echo session.updated but must not re-trigger audio capture or the greeting path. currentSessionConfig is now reset to null at the top of startSession so a stale config from a failed previous session cannot leak into the new one. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): assert wsUrl presence for Gemini proxy connections Without this guard, a missing wsUrl in the hub token response would silently attempt to connect directly to Google with "proxied" as the API key — producing a confusing auth failure instead of a clear error. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): correct Qwen audio formats and default voice DashScope realtime API accepts only 'pcm' for both input and output audio formats. The pcm16/pcm24 values caused session.update rejection before audio capture could start, leaving the Qwen backend unusable. Also updates the default voice from Mia (not in the qwen3-omni-flash- realtime voice list) to Cherry, which is documented as supported. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): close AudioContext on failed voice session start Failed token fetch, microphone denial, or WebSocket error during setup left state.playbackContext open. Each failure path now calls cleanup() before throwing/rejecting, preventing AudioContext leaks on mobile browsers with hard limits on concurrent contexts. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * chore: restore non-voice files to upstream/main state Reverts changes to files that shouldn't differ from upstream: - .gitignore: remove fork-only AGENTS.local.md entry - web/src/App.tsx: restore dual-subscription SSE pattern (scope-aware) - web/src/hooks/useSSE.ts: restore SSEScope/scope parameter - web/src/hooks/useSSE.test.ts: restore (was accidentally deleted) - web/src/lib/appSseSubscriptions.ts: restore (was accidentally deleted) - web/src/lib/appSseSubscriptions.test.ts: restore (was accidentally deleted) - hub/src/sync/syncEngine.ts: restore (off-topic change) * fix(voice): harden Gemini and Qwen WS proxies against client abuse Hub sends HAPI-owned Gemini setup on proxy connect and rejects client setup frames. Qwen proxy always uses QWEN_REALTIME_MODEL instead of a client query parameter. Shared buildGeminiLiveSetupMessage() keeps wire format in one place. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(voice): harden Qwen proxy — hub-owned setup, client frame allowlist Mirror the Gemini proxy security model for Qwen: - Hub sends initial session.update (voice/tools/instructions) on upstream connect so the browser cannot override config fields. - Proxy message() now calls isQwenSafeClientFrame() and closes the connection (1008) if a client session.update touches any field other than 'instructions' (blocks tool/voice/modality overrides). - QwenVoiceSession no longer sends session.update on session.created; it waits for the hub-relayed session.updated and then sends only instruction-only updates for context/proactive content. - Language passed as query param (?language=zh) so hub builds the correct Chinese system prompt without a client-supplied session.update. - buildQwenSessionUpdateMessage() and isQwenSafeClientFrame() added to @hapi/protocol/voice; 9 new unit tests cover filter edge cases. * fix(voice): respect Qwen session.created→session.update protocol ordering DashScope requires session.update to be sent AFTER session.created is received, not immediately on WebSocket open. Previously the hub sent session.update in upstream.onopen, which violated this ordering and risked the config being processed in an uninitialized session context. Add pendingSetupMap to buffer the hub-owned session.update payload. The onmessage handler now relays session.created to the browser first, then immediately sends the pending session.update to DashScope — matching the protocol ordering the old browser-side code used (which waited for session.created before sending session.update). Also remove maxHttpBufferSize from the socket.io Engine config. That setting is unrelated to voice backends; upstream/main had no such limit set and it is not introduced by this PR. * fix(voice): use Realtime tool shape for Qwen session.update (not chat-completions) Qwen Realtime session.update expects tools as flat objects: { type: 'function', name, description, parameters } The previous code used the chat-completions shape: { type: 'function', function: { name, description, parameters } } DashScope may reject session.update or silently ignore tools with the nested shape, causing tool calls to fail at runtime. Fix applied in buildQwenSessionUpdateMessage(); test updated to assert flat shape and that no nested `function` key is present. * fix(voice): update Qwen Realtime model, voice, and endpoint for intl service Live-tested against DashScope international API: - Model: qwen3-omni-flash-realtime → qwen3.5-omni-flash-realtime (previous model ID did not exist on DashScope) - Default voice: Cherry → Tina (confirmed from session.created response on qwen3.5-omni-flash-realtime) - Default WS base: dashscope.aliyuncs.com → dashscope-intl.aliyuncs.com (international accounts use the -intl endpoint; China endpoint rejects international API keys; QWEN_REALTIME_WS_URL env var still overrides) * fix(voice): correct Qwen text injection and generalise language handling Two dogfooding fixes verified against live Qwen Realtime session: sendTextMessage: switch from instruction-injection to conversation.item.create Qwen Realtime requires a user conversation item before response.create. The previous approach (updateInstructions + response.create) produced "input messages do not contain elements with role user" errors. Now sends {type:message, role:user, content:[{type:input_text}]} then response.create. sendContextualUpdate is unchanged (instruction-only, no response trigger). Language handling: replace zh-only branch with buildVoiceLanguageBlock() Previously, only language='zh' added any instruction; all other languages (including English) sent no language block, causing Qwen to drift to Chinese. buildVoiceLanguageBlock() now covers three cases: - 'zh'/'zh-*': existing Chinese block (unchanged) - explicit code ('en','es','fr',...): "Always respond in [Language]" - undefined/auto: "Detect the user's language and maintain it" Applied to buildGeminiLiveConfig, buildQwenSessionUpdateMessage, and the client-side currentInstructions mirror in QwenVoiceSession. Also removes the Gemini hub proxy's zh-only filter, which was discarding explicit language selections other than Chinese. * fix(hub): gate Gemini client frames until upstream setupComplete Hub sends its owned setup on upstream open, then waits for Google's setupComplete acknowledgment before flushing queued client frames. isGeminiSetupCompleteFrame() detects the {"setupComplete":{}} message; message() queues instead of forwarding while pendingMap is live. Addresses the repeated Major finding from bot review on PR #743. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(hub): cap Gemini setup-window pending queue at 1 MiB An authenticated client could flood the queue between upstream.onopen and Google's setupComplete acknowledgment. Add pendingBytesMap tracking and close with 1009 if the budget is exceeded. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(gemini): pass all language codes to hub proxy, not just zh Language selection for French, Spanish, Japanese etc. was silently dropped — only 'zh' was forwarded as a query param. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(voice): expand LANGUAGE_NAMES to cover full ElevenLabs language set Codes like 'no', 'da', 'fi', 'pt-br', 'bg', 'ro', 'cs', 'el', 'ms', 'tl', 'uk', 'hu', 'hr', 'sk' were falling through to raw-code prompts ("Always respond in no"). Now resolve to proper display names. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> --------- Co-authored-by: HAPI <noreply@hapi.run> Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
18bcb522e1 |
feat(web): per-session scratchlist (workbench) panel (#772)
* feat(web): per-session scratchlist (workbench) panel Adds a per-session "scratchlist" panel above the composer for parking notes / drafts / parking-lot ideas that are explicitly held — never auto-sent. This is distinct from the existing queue (QueuedMessagesBar): - Queue = conveyor belt: messages auto-fire once the agent is idle. - Scratchlist = workbench: held until the operator promotes them. The amber accent and "held — not sent" pill make the visual distinction obvious so operators don't mistake one for the other. Features: - Collapsible per-session panel (collapsed by default, persisted in localStorage). - Add (Enter) / delete / reorder (up/down) entries. - Promote-to-composer copies into the composer for editing (entry stays — copy semantics). - Promote-to-queue routes through the existing onSend path so the entry shows up in QueuedMessagesBar; entry is removed only on accepted send. - Entries persist per session under hapi.scratchlist.v1.<sessionId>. - Confirm-on-delete only for entries longer than 100 chars. - Ctrl/Cmd+Shift+S focuses the add-input. - en + zh-CN strings. v1 scope: localStorage-only. Hub-sync deferred to v2 to keep the diff small and reviewable. Test coverage: - web/src/lib/scratchlist.test.ts — 21 tests (storage round-trip, add/delete/reorder/cap, malformed-JSON resilience, confirm threshold). - web/src/components/AssistantChat/ScratchlistPanel.test.tsx — 13 tests (collapse persistence, hydration, add/delete/reorder UI, promote-to-composer copy semantics, promote-to-queue accepted / rejected paths, per-session isolation). Closes #11 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): block focus into collapsed panel via inert Upstream review (tiann/hapi#772, codex bot) flagged that the collapsed scratchlist body was visually hidden via CSS only - the textarea and action buttons stayed mounted, focusable, and clickable while their ancestor was aria-hidden. Tab into invisible controls + a hidden subtree with focusable descendants is an a11y violation. Apply `inert` to the inner content, gated on the collapsed state. This removes the subtree from the focus, pointer, and accessibility trees while keeping the grid-template-rows expand animation intact (no conditional remount, so the open/close transition still runs). Add a regression test that asserts `inert` is present while collapsed and removed (or empty) while expanded, so a future revert of the fix trips immediately. Co-authored-by: Cursor <cursoragent@cursor.com> * test(scratchlist): add Playwright e2e + isolated fixture page The unit suite under jsdom can't verify the parts of the scratchlist that actually live in the browser: - `inert` blocks focus (jsdom ignores `inert`) - the grid-template-rows collapse animation - localStorage surviving a full page reload - per-session keying surviving cross-route navigation - Ctrl/Cmd+Shift+S firing the global expand+focus shortcut Add a Playwright config + spec that drives a real Chromium against a new Vite-served fixture (`web/e2e-fixtures/scratchlist-fixture.html`). The fixture mounts the production `ScratchlistPanel` in isolation inside an `I18nProvider` and exposes the promote callbacks on `window.__scratchlistE2E` so the spec can assert that promote-to- composer and promote-to-queue receive the right text without having to spin up the hub, auth, or socket layer. Nine specs cover: 1. starts collapsed, toggles 2. collapsed inner is `inert` and refuses focus / pointer 3. add: entry appears, draft clears, count updates 4. persistence across full page reload 5. promote-to-composer fires callback (entry stays - copy semantics) 6. promote-to-queue success path (entry removed) 7. promote-to-queue failure path (entry retained for retry) 8. Ctrl+Shift+S expands + focuses input 9. per-session isolation across navigation Wires `bun run test:e2e` and `test:e2e:ui` at the repo root and documents the harness in `web/README.md`. Bumps `playwright` 1.49.1 -> 1.60.0 alongside the new `@playwright/test` dep so the bundled chromium-headless-shell-1223 (Chrome 148) is used; the older 131 binary SIGTRAPs on this kernel during launch. Adds `test-results/` and `playwright-report/` to `.gitignore`. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(scratchlist): key host by session.id to prevent cross-session leak Upstream review (tiann/hapi#772, codex bot follow-up) flagged a state leak across same-route session switches. ScratchlistPanel reads `sessionId` once via `useState(() => readScratchlist(sessionId))` and rehydrates in a `useEffect`. SessionChat stays mounted when the operator switches sessions on the same `/sessions/$sessionId` route, so the panel sees a new `sessionId` prop without unmounting. Effect order during the prop change: 1. render with sessionId=B but stale entries=[A's items] 2. rehydrate effect: setEntries(read(B)) -> queues correction 3. persist effect (deps [sessionId, entries] both changed): persistScratchlist(B, [A's items]) -> writes A into B 4. re-render with sessionId=B, entries=B's items 5. persist effect: persistScratchlist(B, B's items) -> overwrites the bug write The bug is transient (step 3's write is corrected by step 5) but real: any read between steps 3 and 5 (another tab, a SW prefetch, manual inspection) sees A's data under B's key. Fix is one line: `key={props.session.id}` on `<ScratchlistHost>`. React unmounts and remounts the host when the key changes, so the new mount's useState initializer reads B's storage from scratch and never touches B's key with A's data. This is the React-canonical "reset state on prop change" pattern; cleaner than chasing the race inside the panel. Add an e2e regression test that: - installs a `localStorage.setItem` spy in `addInitScript` - mounts the fixture under session A and adds an entry - clears the spy, then switches to session B in-place via `window.__scratchlistE2E.setSessionId('leak-B')` (no page reload) - asserts no recorded write to `hapi.scratchlist.v1.leak-B` contained A's text (catches the transient corrupting write deterministically, before the correction overwrites it) - round-trips back to A to confirm A's storage is intact The fixture grows a `?key=0` mode that drops the host's `key=` prop. Verified red/green: with `key=0` the regression test fails on the spy-detected corrupting write; with the fix in place (default), all 10 e2e specs pass. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
hangerandGitHub
|
bffbe1b861 |
fix(web): add close button to dialog so modals are dismissable on mobile (#792)
* fix(web): add close button to dialog so modals are dismissable on mobile The shared DialogContent had no close affordance — desktop users could press Escape or click the overlay, but on mobile (no Escape key, dialog spans calc(100vw-24px) leaving almost no tappable overlay) there was no way to dismiss it. Add a DialogPrimitive.Close X button in the top-right, fixing every dialog that uses this component at once. * fix(web): reserve header space for dialog close button Address review feedback: the absolutely-positioned close button overlaps the top-right of every dialog. Long/breaking titles (e.g. DiffView's break-all filename) could wrap underneath the 32px tap target. Add pr-12 to DialogHeader rather than padding DialogContent globally, so the title row clears the button while body content (code blocks, diffs) keeps full width. * fix(web): localize dialog close button aria-label Use the existing button.close locale string instead of a hardcoded "Close" so screen-reader users get the label in their language (zh-CN: 关闭). |
||
|
|
f9ef3a4489 |
feat(codex): import local Codex sessions into Hapi (#796)
* local: add Codex Desktop session sync controls * feat(codex): import local Codex sessions into Hapi --------- Co-authored-by: Codex Local <codex-local@example.invalid> |
||
|
SSU-WEI HUANGandGitHub
|
39fba5292c | fix: add Telegram notification context (#768) | ||
|
|
bd13fac7ae |
fix(claude): apply mid-turn permission mode changes to canCallTool (#764)
`PermissionHandler` stored its own `permissionMode` field and only updated it inside `handleModeChange`, which is called when a new batch is pulled from the queue. The `SetSessionConfig` RPC (web dropdown changes) updates `runClaude.ts`'s `currentPermissionMode` and the session keepalive metadata, but never reaches the handler — so switching to Yolo mid-turn left `canCallTool` checking the stale mode and still prompting for approval. Closes #735. Drop the stored field and read live from `session.getPermissionMode()`, mirroring how the OpenCode permission handler already works. Override `Session.getPermissionMode()` in `claude/session.ts` to return the Claude-narrow `PermissionMode`, sound because the matching `setPermissionMode` setter only accepts that subset. via [HAPI](https://hapi.run) Co-authored-by: HAPI <noreply@hapi.run> |
||
|
junesandGitHub
|
30564601ed |
fix(cli,web): hide Windows spawn windows and show queued attachments (#765)
* fix(cli,web): hide Windows spawn windows and show queued attachments * fix(web): preserve attachment-only queued edit text |
||
|
weishu
|
4aee1e4f84 | Release version 0.19.0 | ||
|
|
d78cf4b171 |
fix(cli): Fix Codex CLI execution issue in PowerShell with Hapi Codex (#763)
* fix(cli): fixed an issue where the codex cli failed to run successfully when using hapi codex in powershell * fix(cli): Fixes the issue of Windows Codex npm shim bypassing the launcher --------- Co-authored-by: xhd902 <xuhang@infypower.cn> |
||
|
weishu
|
ec1ab23e6c | Reduce automatic title update prompts | ||
|
|
449cf6af0a |
feat(cli): wire Cursor /summarize and /clear slash builtins (#747)
* feat(cursor): wire /summarize and /clear slash builtins for remote sessions Seed cursor builtins for web autocomplete, parse summarize/clear in cursorRemoteLauncher (pass-through to agent -p; reject /clear with args). Fixes tiann/hapi#738 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(cursor): isolate slash commands before message queue batching Parse summarize/clear at enqueue time (runCursor) with pushIsolateAndClear so waitForMessagesAndGetAsString never merges a slash with the next prompt. Adds queue policy tests for invalid /clear + following message. Addresses PR #747 review. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(cursor): preserve pending messages when isolating slash commands pushIsolateAndClear() wipes the entire queue, so a normal prompt queued before /summarize or /clear would be silently dropped. Add pushIsolated() - isolation without clearing - and route Cursor slash commands through it instead. Adds queue tests covering the preserve-then-isolate path. Addresses PR #747 review. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
02d4e93178 |
fix(acp): drop mid-stream usage emit; OpenCode only sends usage_update at end-of-turn (#760)
PR #756 added a mid-turn emit in captureUsageUpdate to surface live context usage via the web status bar. Testing against OpenCode 1.15.11 on a real session shows OpenCode emits a single usage_update per turn, within ~1ms of session/prompt resolving — never during streaming. That makes the mid-turn path dead code for OpenCode (and for any other ACP agent that follows the same pattern). It also persists a useless inputTokens:0/outputTokens:0 token_count message that gets immediately overwritten by the finalize emit, churning the session history. Drop the mid-stream emit and the activeOnUpdate plumbing it required. Keep the finalize fallback for agents that don't return a usage block on session/prompt (slash-handled turns, errored turns). The persistent "live" counter requires the agent to emit usage_update during streaming; filed upstream against anomalyco/opencode. Refs #750 via [HAPI](https://hapi.run) Co-authored-by: HAPI <noreply@hapi.run> |
||
|
|
cc4025abdb |
fix(web,hub): queued bar SSE + never-started inactive resume (#761)
* fix(web): apply messages-consumed on global SSE connection The global all-sessions SSE subscription returned early on message-stream events without updating the message-window store. When session-scoped SSE was reconnecting or the user had another session selected, messages-consumed never cleared the queued bar even though the hub had stamped invoked_at. Also harden mergeMessages so a stale invokedAt:null snapshot cannot clobber an existing ack timestamp. Fixes tiann/hapi#758 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web,hub): resume never-started inactive sessions on first send Hub fresh-spawns when inactive session has path but no agent thread id and zero messages. Web guards resume, updates inactive banner copy, and surfaces resume_unavailable before POST /resume when resume is impossible. Fixes tiann/hapi#759 Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): scope sessionResume guard to current flavor only Hub `resolveAgentResumeId` only honors the metadata.flavor's id; the web guard was falling back across all flavors so a cursor session with a stale codexSessionId still tried to resume and 409'd. Mirror the hub switch and default to claude when flavor is unknown. Addresses HAPI Bot review on tiann/hapi#761. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): allow claude session resume via hub message-id recovery Hub `resolveAgentResumeId` falls back to `recoverClaudeSessionIdFromMessages` on the claude branch when `metadata.claudeSessionId` is absent, so the web guard must not block inactive claude sessions that have stored messages but no metadata id. Other flavors have no such recovery path and stay rejected. Addresses second HAPI Bot review thread on tiann/hapi#761 (`web/src/lib/sessionResume.ts:41`). Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
SSU-WEI HUANGandGitHub
|
df35a8c523 | fix(test): isolate integration tests from production hub via temp hub globalSetup (#734) | ||
|
SSU-WEI HUANGandGitHub
|
994a820e43 | fix(opencode): surface ACP context usage live to web status bar (#756) | ||
|
junesandGitHub
|
a1d144d290 | fix(web): handle legacy session summaries (#751) | ||
|
SSU-WEI HUANGandGitHub
|
5b797bb95d | feat(opencode): slash command support (#671) (#753) | ||
|
junesandGitHub
|
31dd4353d4 | fix(cli): replace existing runner on start (#754) | ||
|
SSU-WEI HUANGandGitHub
|
c09bbaed3d | fix(codex): render /help and /status as markdown so web shows line breaks (#755) | ||
|
|
4b24528362 |
fix(acp): flush straggler chunks promptly after session/prompt returns (#730)
* fix(acp): flush straggler chunks promptly after session/prompt returns After session/prompt returns, HAPI drains buffered agentMessageChunk text and marks the turn complete, but leaves the message handler alive. Models with long streaming tails (DeepSeek, GPT-5.5) continue to push chunks after that drain, causing text to accumulate in the buffer and only appear when the next user prompt triggers the pre-prompt drain — showing up in the wrong turn with broken markdown. Start a 50ms interval timer after the post-prompt drain that keeps calling drainBuffers() on the live handler for up to 6 seconds, so straggler chunks are emitted within one poll tick instead of waiting for the next prompt. The timer is cancelled when the next prompt starts (pre-prompt drain replaces the handler) or on disconnect. Fixes #609. Also applies to Gemini and Kimi which share the same AcpSdkBackend code path. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(acp): gate next turn's handler swap on previous turn's late drain Addresses the github-actions review on #730: the fire-and-forget late-flush timer let `prompt()` resolve while stragglers were still possibly arriving, so a rapid follow-up prompt could either drop those chunks (during the old null-handler gap) or leak them into the new turn's onUpdate. Pre-prompt phase now keeps the previous turn's handler alive across the quiet wait (bounded by LATE_FLUSH_WINDOW_MS) and swaps in a single phase immediately before sending the new session/prompt. The post-prompt late flush timer is unchanged — it still emits idle-window stragglers promptly without delaying the ready signal or `setModel` / `setConfigOption`. Adds three regression tests: late-chunk flushing within the window, pre-prompt straggler attribution to the previous turn's onUpdate, and disconnect cancelling the timer. Removes now-unused PRE_PROMPT_UPDATE_DRAIN_TIMEOUT_MS. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(acp): await late drain so ready never fires before stragglers emit Follow-up to bot's repeated MAJOR on #730: even with the pre-prompt gate, the fire-and-forget late-flush timer let `prompt()` resolve before slow tails finished, so the launcher's `ready` signal (and any user follow-up queued against it) raced with text still being emitted to the current turn's onUpdate. Replace the setInterval timer with a synchronous `drainLateBuffers()` awaited in `prompt()`'s finally before turn_complete is sent. It polls drainBuffers every LATE_FLUSH_INTERVAL_MS so the UI keeps streaming smoothly during the wait, and exits early once the model has been quiet for LATE_FLUSH_QUIET_PERIOD_MS (250 ms — adds negligible latency to fast models like Claude whose tail is typically <100 ms) or the LATE_FLUSH_WINDOW_MS upper bound (6 s) elapses. Side effects: - Restore PRE_PROMPT_UPDATE_DRAIN_TIMEOUT_MS (1200 ms): the pre-prompt drain is now just a safety net since the post-prompt wait guarantees the previous turn is quiet by the time the next prompt starts. - Drop the `lateFlushTimer` field, `startLateFlushTimer`, `stopLateFlushTimer`, and their disconnect/pre-prompt cleanup calls. - Update the "emits straggler chunks" test to assert ordering before turn_complete, and add a fast-path test confirming the drain exits promptly when the model is quiet. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * fix(acp): anchor late-flush quiet window to entry, not stale lastSessionUpdateAt Bot's third MAJOR on #730: drainLateBuffers() compared elapsed time against lastSessionUpdateAt, which can already be older than LATE_FLUSH_QUIET_PERIOD_MS by the time the method starts — e.g. when the model emits chunks early in the turn, pauses, then sends stopReason. In that case the first loop iteration sees a stale "quiet" reading and returns immediately, missing any straggler that arrives just after session/prompt resolves; the chunk then sits in the buffer until the next prompt's pre-prompt drain. Anchor the quiet check to max(lastSessionUpdateAt, entry time) so we always observe at least one quiet period from method entry regardless of when the last chunk was. Adds a regression test that fires a chunk early, awaits a 200ms pause, schedules a post-resolution straggler, and asserts it lands before turn_complete. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> * docs(acp): correct LATE_FLUSH_QUIET_PERIOD_MS comment after entry-anchor fix The previous note claimed the 250ms quiet check "exits early for fast models, adding negligible latency". That was true before commit 512d6a4 when the check compared against lastSessionUpdateAt; with the entry-time anchor, drainLateBuffers always observes at least one full quiet period. Document that this minimum wait is the price of catching post-resolution stragglers from paused-mid-turn models. via [HAPI](https://hapi.run) Co-Authored-By: HAPI <noreply@hapi.run> --------- Co-authored-by: HAPI <noreply@hapi.run> |
||
|
junesandGitHub
|
9e17953c11 | fix(cli): resolve Windows Claude npm shim (#739) | ||
|
|
3cc03eb9da |
fix(hub): skip redundant set-session-config RPC on resume (#740)
resumeSession already passes permissionMode to spawnSession. The follow-up applySessionConfig raced session-alive (handler not registered yet) and returned resume_failed after hub restart. Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
|
c58e8cea9e |
fix(cursor): persist resume id early and return 409 for resume_unavailable (#745)
Remote cursor launcher now mirrors local launcher by writing cursorSessionId to hub metadata as soon as --resume is known, before the agent init event. POST /sessions/:id/resume maps resume_unavailable to 409 with clearer guidance. Fixes tiann/hapi#744 Co-authored-by: Cursor <cursoragent@cursor.com> |
||
|
hangerandGitHub
|
6200ae1370 |
feat(web): align Claude effort options with Claude Code --effort levels (#731)
The Claude effort selector (New Session config + in-session composer) only offered auto/medium/high/max, missing `low` and `xhigh` — yet `claude --effort` actually accepts low/medium/high/xhigh/max. Add the two missing levels in both places so the selector faithfully mirrors the CLI. Extract the level list + labels into one shared constant (@hapi/protocol: shared/src/effort.ts, mirroring CLAUDE_MODEL_PRESETS) so the two UIs derive from a single source and can't drift again. No backend change: the effort string is free-form end-to-end through to the --effort flag. ultracode is intentionally excluded — it is a TUI-only /effort session setting, not an --effort value (the CLI rejects `--effort ultracode`). |