Merges FN-3139 planning comment infrastructure (comment inputs in mission/milestone-slice/planning modals, comment display in conversation history), FN-3119 Quick Chat FAB with slash-triggered skill menu and plugin integration, FN-3117 plugin dashboard views and chat session icons, and FN-3066 plugi
Fusion-Task-Id: FN-3139
The vitest child-process guard wrapped exec/execFile without preserving the
`[util.promisify.custom]` symbol, so awaited `execAsync` resolved to a raw
stdout string instead of `{stdout, stderr}`. That single regression cascaded
through ~60 "failing" tests across cli, core, engine, and dashboard whose
production code was actually correct. Also relax the AI-CLI blocklist for
cheap introspection (--version/--help/which …), give SIGTERM'd subprocesses a
brief grace period before being flagged as "left running", fix a few real
test-side bugs uncovered along the way (executor mock step transitions, iOS
last-resort keyboard path, mission SSE replay tests racing with the real AI
agent), and convert dashboard route tests' dynamic `await import("../server.js")`
to static imports so first-test timings drop from 2–5s to <200ms.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Subtask, mission-interview, and milestone/slice-interview sessions could pin
their `generating` state forever when the underlying provider stream stalled
silently or a tool call hung. Wrap each `agent.session.prompt()` in a new
GenerationGuard helper (per-session AbortController + timer) so a stuck turn
becomes a bounded error users can retry. Adds matching `stop*Generation`
exports and threads abort through cleanup so dismissing a modal cancels the
in-flight call instead of leaking it.
Also closes the gh-cli tool hang vector: `runGhAsync` / `runGhJsonAsync` now
accept `{ signal, timeoutMs }` (default 30s). Github-touching extension tools
forward the AI tool's signal so an aborted agent kills the `gh` child instead
of orphaning it.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The dashboard modules used a variable-specifier dynamic import
(`const m = "@fusion/engine"; await import(m)`) to defeat bundler static
analysis. tsup honored that and left the dynamic import in dist/bin.js,
so the published `@runfusion/fusion` package failed at runtime with
"createFnAgent2 is not a function" — `@fusion/engine` isn't on npm and
the silent catch set the binding to undefined. Replaces the trick with
static imports across planning, chat, subtask-breakdown, mission-interview,
agent-generation, ai-refine, roadmap-suggestions, milestone-slice-interview,
and routes. Core can't statically import engine (cycle), so it now exposes
setCreateFnAgent and engine wires itself in at module load. Documents the
pattern in AGENTS.md.
FixesRunfusion/Fusion#9.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Migrate mission, milestone-slice, and subtask breakdown flows to shared ai-session-diagnostics logging
- Add a planning-flow guardrail test that fails on raw console.* diagnostics across planning modules
- Update mission and subtask breakdown tests to use structured diagnostics sink hooks instead of console spies
- Keep backward-compatible diagnostics test hooks and reset injected sinks to avoid cross-test leakage
- Rename engine export and call sites to use createFnAgent consistently across runtime flows
- Update core lazy engine loader and dashboard agent-generation/planning/chat paths to reference createFnAgent
- Refresh affected unit and integration tests, including renaming pi-create-kb-agent.test.ts to pi-create-fn-agent.test.ts
- Update AGENTS.md documentation references to match the new createFnAgent name
- Add defensive cleanup for in-memory task tracking when tasks move state or agents pause
- Improve stuck-detection and self-healing flow to reduce leaked state and missed recovery paths
- Add logging around previously swallowed errors and tighten executor cleanup behavior
- Expand restart and self-healing reliability tests to cover regression scenarios
- Add projectId parameter to refineText API and TaskForm for proper scoping
- Update mission routes to require projectId and add regression tests
- Add mission interview fallback to mission-level context when target not found
- Add AI refine route scoping tests for project isolation
- Update component tests to expect projectId argument
- Update prompt-keys test counts for new system prompts
- Bump SQLite schema to v19 with ai_sessions lock columns and lock index, and align migration coverage in core DB tests
- Extend AiSessionStore with acquire/release/force lock APIs, stale lock cleanup, and lock metadata in ai_session update summaries
- Enforce lock checks on planning, subtask, and mission interview mutation routes with 409 conflict responses while keeping stream reads unaffected
- Add frontend tab identity + useSessionLock hook and wire Planning, Subtask, and Mission modals to pass tabId, show lock overlay, and support Take Control
- Expand dashboard route/e2e and modal tests to validate lock enforcement, lock handoff, and lock-aware session reentry behavior
- Keep errored AI sessions retryable in the session store and add backend retry handlers for planning, subtask breakdown, and mission interview flows
- Add retry API routes and dashboard API client helpers for planning, subtask, and mission interview session retries
- Update PlanningModeModal, SubtaskBreakdownModal, MissionInterviewModal, and background session handling to show error states with retry/cancel UX
- Expand unit and integration tests across store, services, routes, and modal components to cover retry success and failure paths
- Persist per-turn thinking output for planning and mission interview sessions, including history serialization and recovery-safe state fields
- Add shared conversation history parsing/API types and a reusable timeline component with formatted answers and expandable AI reasoning
- Restore and render conversation history across Planning Mode, Mission Interview, and Subtask Breakdown modals during resume and live question flow
- Harden response submission with nullable-question guards and expand unit/e2e coverage for history rendering and persistence behavior
- Add recoverable-session querying in AiSessionStore and cover it with targeted store tests
- Persist resume context (ip, initial plan, mission metadata) and rebuild planning/subtask/mission sessions from SQLite rows
- Rehydrate recoverable sessions at server startup and resume planning/mission interviews by recreating agents with replayed conversation context
- Update API flows to pass project root context and clean in-memory sessions when persisted sessions are deleted, with comprehensive regression tests
- Add aiSessionTtlMs and aiSessionCleanupIntervalMs project settings with defaults and bounds for cleanup scheduling
- Extend AiSessionStore cleanup to expire stale in-progress sessions, emit deletion events, and support start/stop scheduled cleanup loops
- Wire server startup/shutdown to load cleanup settings and manage the scheduled ai_sessions sweep lifecycle
- Align planning, subtask breakdown, and mission interview in-memory session retention to a 7-day TTL with shared deletion-driven cleanup
- Update schema/tests/docs for migration v15 ai_sessions indexing and end-to-end cleanup/TTL behavior stability
- Add a shared SSE event buffer utility and wire it into planning, subtask, mission interview, and task stream routes
- Support Last-Event-ID replay semantics and robust event serialization so clients can recover missed stream events
- Add a resilient client-side reconnect wrapper and surface reconnecting state in planning, subtask breakdown, and mission interview modals
- Expand test coverage across API routes, SSE buffering behavior, reconnect handling, and mission/planning stream flows
Introduces ai_sessions table (schema v9), AiSessionStore, and background
session support for mission interviews, planning, and subtask breakdown.
Adds BackgroundTasksIndicator component, SSE-based progress streaming,
and dashboard API routes for session management.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add a conversational AI interview flow to the missions screen that
interviews users to break down missions into milestones, slices, and
features with verification criteria at every level.
Backend: Replace stubbed interview logic with real AI agent integration
using createKbAgent, SSE streaming, and robust JSON parsing. Wire up
all 5 placeholder mission interview route endpoints (start, respond,
cancel, stream, create-mission).
Frontend: New MissionInterviewModal component with conversational UI,
thinking output display, question renderers, and hierarchical plan
review/edit. Integrated into MissionManager with "Plan with AI" button.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add openrouterModelSync setting to eagerly fetch OpenRouter model catalog at
dashboard startup so the model picker shows all available models
- Add toggle in Settings → Models to disable the sync
- Add mission-interview session management module for AI-guided mission specs
- Include tests for the sync opt-out behavior
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
