Merges the final step for FN-3309, adding a changeset to document the removal of pi install references. The changeset is the only file changed, completing the feature work.
Fusion-Task-Id: FN-3309
Adds a patch changeset for FN-3290 documenting the modal keyboard isolation fix, ensuring the change is captured in the release changelog.
Fusion-Task-Id: FN-3290
The merge restores Claude usage tracking by introducing a Proxy-based auth storage with a fallback resolver that falls back to `models.json` API keys when the primary auth store lacks credentials. It also adds planning improvements with corresponding tests and a context limit detector enhancement, a
Fusion-Task-Id: FN-3305
Merges FN-2999 research hardening (idempotent cancel/retry routes, aligned SSE event wiring, and cleaned status handling in the core research store and orchestrator) plus UI improvements to AgentDetailView header actions and planning disclosure UX in the modal, with a CSS token fallback fix in Scrip
Fusion-Task-Id: FN-2999
This merge adds Docker connectivity management to Fusion (FN-3109), introducing a Docker client service in `@fusion/core` with typed interfaces for daemon connection configuration (host, socket path, TLS). The dashboard gains a `DockerTargetSelector` component for choosing between local socket and T
Fusion-Task-Id: FN-3109
Merged FN-3207 and FN-3242: streamlined the memory backend by removing the runtime side-load pattern in `project-memory.ts` (simplified from 25+ lines), added regression tests across core, engine, and CLI bundle to catch the import issue at build time, and updated ChatView with a CSS fix for file me
Fusion-Task-Id: FN-3207
This merge adds a durable insight lifecycle with bounded run executor to the core store layer, along with supporting database schema changes and comprehensive test coverage. It also restores type compatibility for list view detail callbacks and fixes QuickChatFAB CSS scope issues. Documentation acro
Fusion-Task-Id: FN-3243
The dashboard re-instantiates AgentStore on every /api/agents request, and
each instance was opening a fresh Database — re-running schema migrations
and PRAGMA integrity_check on the full file. On a multi-hundred-MB DB that
was 3-7s per request and leaked file handles. Sharing a cached Database
keyed by rootDir reduces it to a one-time process cost.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Group the planning model selector and depth controls under a collapsible
"Advanced planning settings" disclosure (defaultOpen) in the Planning Mode
modal. Adds a defaultOpen prop to OnboardingDisclosure and tightens the
node.capabilities readonly typing in NodeDetailModal.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Speeds up task-list load and interaction for projects with 100+ tasks.
- Migration 59 adds idxTasksColumn and idxTasksUpdatedAt; listTasks()
filters by "column" on every board load and SSE/refresh paths sort by
updatedAt, so each query was previously a full table scan plus a temp
B-tree sort.
- Debounce handleEmbeddedOpenDetail in ListView so rapid keyboard/mouse
navigation no longer fires a heavy /tasks/:id (log + comments) per
selection; stale-target requests short-circuit.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Closes the loop on the blur/close summarize work. Previously the sidebar
always rendered `session.preview` for drafts, so the model call fired by
blur/close updated `session.title` to a value the user never saw, and
then `startExistingSession` summarized again on Start Planning even when
blur had already produced a title for the exact same text.
The sidebar now shows the summarized title once available and falls
back to the inputPayload preview only while the title is still the
placeholder. summarizeDraftTitle persists the plan text it summarized
from via a new `markDraftSummarized` store method, and updateDraft
preserves that marker only while the persisted text still matches it
(text edits invalidate it automatically). startExistingSession reads
the marker and skips its own summarize when the persisted plan equals
what was summarized last, eliminating the redundant model call.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
pi-coding-agent sets Skill.name to the parent directory (e.g. 'web-research'),
while Fusion uses two-segment names everywhere (e.g. 'web-research/SKILL.md')
from extractSkillName(), normalizeAgentSkills(), and toggleExecutionSkill().
The previous fix (f2afc7f0) correctly switched matching from skill.filePath to
skill.name, but that only works when both sides use the same format. Since they
don't, all pattern/requested-name comparisons still failed, producing the
spurious 'not found in discovered skills' warnings.
Fix: add bareSkillName() helper that strips the /SKILL.md suffix before
comparison. Applied to all five comparison points in skill-resolver.ts:
- skillNameMatches() (pattern filtering)
- requestedSkillNames set lookup (name filtering)
- hasDiscoveredMatch() (configured-pattern diagnostic)
- discoveredBareNamesLower (requested-name diagnostic)
- excluded-path discovery check
The planning modal's draft sync only updated SQLite, never the in-memory
session, so Start Planning ran the agent against whatever fragment was
typed before the first 500ms debounce and silently dropped everything
after. Drafts that survived a backend restart couldn't be started at
all (lazy rebuild was missing for the start path), reopened drafts left
the textarea empty, and the model override the user picked at create
time was forgotten on reopen — Start Planning would silently use the
project default instead.
This change rehydrates initialPlan from SQLite in startExistingSession,
flushes the latest text from the request body before reading it back,
lazy-rebuilds the in-memory session from the persisted row when needed,
and round-trips the model override through inputPayload so reopen
restores it into modal state. Sidebar drafts now render a derived
preview (from inputPayload, never persisted as title) so multiple
drafts are distinguishable while editing, and summarizeDraftTitle fires
on textarea blur and modal close — gated on status rather than title
content so blur-then-edit still refreshes rather than locking to the
first snapshot.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
`TriageProcessor.finalizeApprovedTask` (added in FN-3056) called
`store.updateTask({title})` while the task was still in column='triage',
which triggered a pre-existing regen path in `TaskStore.updateTask` that
overwrote the agent's just-written specification with the bootstrap stub
(`# {id}: {title}\n\n{description}\n`). Tasks shipped to `todo` (and
through to `done`) with empty 70–200 byte specs while the executor only
saw the original one-line user description. The same regen path also
silently dropped `## Review Level` / `## Frontend UX Criteria` and any
section outside a fixed whitelist whenever a non-triage task's title or
description was edited.
Replaces the regen with wrapper-shape-exact stub detection (compare to
the bytes `createTask` would have written for the pre-update title and
description) plus surgical edits for real specs: title changes splice
only the leading `# ...` heading, description changes rewrite only the
body of `## Mission`, and every other section is preserved verbatim.
`finalizeApprovedTask` now applies the prompt-declared title after
`moveTask("todo")` as defense in depth. New regression tests cover real
specs surviving title sync, long bootstrap stubs, stubs whose body
contains `##` markdown or `**Created:**` text, and the end-to-end
triage finalize sequence on a real `TaskStore`.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Merged the three-step FN-3239 feature: auto-installing the dependency graph plugin on first run across the `daemon`, `dashboard`, and `serve` commands, with the core logic centralized in `bundled-plugin-install.ts`, bundling configuration in `tsup.config.ts`, and a documentation file for the default
Fusion-Task-Id: FN-3239
Merges FN-3008 to add a "fallback-used" notification system: the engine now emits events when AI model fallbacks are triggered, dispatches notifications via ntfy/webhook providers, surfaces a session banner in the dashboard, and exposes a settings toggle to enable or disable these alerts.
Fusion-Task-Id: FN-3008
`git diff --no-index` exits 1 when files differ — that's the success case for synthetic untracked-file diffs, not an error. Use spawn directly so we can resolve on exit 0 or 1, independent of how callers (or test mocks) wrap execFile / promisify.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Refreshes the built-in model catalog feeding ModelRegistry with the
latest entries upstream pi-ai generates from models.dev (Anthropic,
OpenAI, Codex, Bedrock, etc.). No Fusion-side API changes; upgrades
applied in cli, dashboard, and engine package.json plus lockfile.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two sync hot paths were stalling the dashboard event loop on a periodic
timer:
1. `pgrep -f vitest` ran via `execSync` in `getVitestProcessIds`
(`/api/system-stats`, `/api/kill-vitest`) and `killVitestProcesses`
(TUI memory-pressure check). On a busy machine pgrep walking the
process table can take 100ms+; execSync blocks the entire Node event
loop for that duration, so every concurrent dashboard request hangs
while pgrep runs. The TUI variant fired on every memory-pressure tick
(~2s when over threshold), the dashboard variant fired on every
system-stats poll (5s while the modal is open). Both now use execFile
with a callback wrapped in a Promise.
2. `discoverDashboardPiExtensions` (called from 3 /api/settings/pi-
extensions routes) did 6+ blocking existsSync/readFileSync calls per
invocation across legacy and fusion settings paths. Converted to
fs.promises.readFile/access and parallelized via Promise.all.
Behavior preserved:
- TUI memory-pressure detection still works (sync os.totalmem path
unchanged); auto-kill still fires on threshold breach.
- The `lastAutoKillAt` 30s re-fire gate is set before the async kill
starts, so concurrent ticks can't trigger duplicate kills.
- system-stats still polls every 5s while the modal is open and still
returns vitestProcessCount.
- All 7 system-stats / kill-vitest tests pass; all 6 settings/pi-
extensions tests pass. Test mocks updated for the (err, stdout, stderr)
callback signature.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three merger fallback commit paths (auto-resolve-all-conflicts,
-X theirs/ours side strategy, AI-agent-didn't-commit) hard-coded
`feat(FN-XXXX): merge fusion/fn-xxxx` as the subject and never used
the AI subject summarizer. Route them through buildDeterministicMergeMessage
so they pick up aiSubject when available.
When the AI subject summarizer returns null, derive the subject from
the branch's first step commit (with conventional-commit prefix
stripped, plus `(+N more)` for multi-commit branches) instead of the
bare `merge <branch>` template.
Bump DEFAULT_COMMIT_SUBJECT_TIMEOUT_MS 15s → 30s so slow-first-token
providers complete instead of silently falling back.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge delivers the full draft planning feature (FN-3106) — API routes for planning subtasks, the AiSessionStore backend for draft sessions, a draft planning API client, auto-creation of planning drafts from user input, and corresponding UI polish in PlanningModeModal with accessibility-focused
Fusion-Task-Id: FN-3106
This merge brings FN-3155's plugin `createAiSession` API (types, DI hooks, engine adapter, context wiring, docs, and tests), FN-3056's task title sanitization, and FN-3129's tokenized footer and mobile initialization for MissionManager. It also adds CentralCore Docker node management, a new AddNodeM
Fusion-Task-Id: FN-3155
Aligns updateSchedule validation with createSchedule so step-based
automations (auto-summarize, memory dreams) can sync at startup
without hitting "Command cannot be empty".
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge ships several feature and infrastructure improvements across the codebase. Task title validation is strengthened in triage with stricter rejection of malformed titles and preference for prompt-declared titles (FN-3056), while task creation now preserves priority settings (FN-3210). The Mi
Fusion-Task-Id: FN-3056
On high-core dev machines (e.g. 28-core M-series), per-package vitest
defaulted to cpus().length - 1 workers (27), and `pnpm test` ran 4
workspace packages concurrently — easily 100+ vitest threads per sweep.
When the dashboard had agents running tests, 2+ concurrent sweeps would
saturate CPU and the UI became sluggish.
- Cap defaultMaxWorkers to min(6, cpus()-1) in cli/dashboard/desktop/
mobile/plugin-sdk vitest configs (engine and core were already capped)
- Lower root `pnpm test` workspace-concurrency 4 → 2
- VITEST_MAX_WORKERS override still respected for explicit fast runs
Worst-case fan-out drops from ~108 workers to ~12 per `pnpm test`.
CI runners with fewer cores are unaffected (cap doesn't bind).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The nested action row kept "Create Single Task" and "Break into Tasks"
side-by-side on small screens, pushing the primary button offscreen.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge implements a "preserve progress" option for task resets across the system. FN-3185 adds a `preserveProgress` flag to `moveTask` that keeps status/history when resetting tasks back to `todo`, with required explicit confirmation dialogs to prevent accidental resets. The feature is wired thr
Fusion-Task-Id: FN-3185
Merged FN-3076 introducing an auto completion-doc mode that automates task completion documentation. The feature adds a new setting to the settings schema and types, surfaces it in the dashboard Settings UI, provides triage-stage guidance to suggest completion documentation, and is documented in the
Fusion-Task-Id: FN-3076
This merge lands v0.15.0, bringing a major plugin system overhaul including a new dependency graph plugin with SQLite-backed storage, Quick Chat FAB with slash-triggered skill menu, and expanded plugin dashboard views. The AgentsView tree mode was fully removed with its hook, styles, and expansion s
Fusion-Task-Id: FN-3121
- Add plugin dashboard view registration and hosting across core, dashboard routes, and plugin SDK exports
- Integrate plugin-provided views into app navigation, mobile/header UI, and view state hooks with coverage
- Add fusion-plugin-dependency-graph example plugin with persisted storage, dashboard view UI, and manifest wiring
- Update plugin authoring and architecture docs for dashboard view extension points
- Add a changeset for @runfusion/fusion covering plugin dashboard view support
Fusion-Task-Id: FN-3079
This merge brings FN-3173's SQLite stability improvements: WAL tuning pragmas for better concurrency, periodic integrity checks with self-healing recovery, and batched agent log writes to reduce I/O overhead. It also includes a new cron-runner for scheduled maintenance tasks, TUI mouse wheel scrolli
Fusion-Task-Id: FN-3173
Enable xterm SGR mouse reporting in the dashboard TUI and dispatch
wheel events to the focused pane: task detail logs, Git lists
(commits/branches/worktrees), and Files view (tree or preview).
Mouse mode is enabled after Ink mounts so the leading ESC of wheel
reports never arrives alone — avoiding spurious Esc keypresses. We
omit motion-tracking modes so terminals still own drag gestures and
Shift+drag native text selection keeps working.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two follow-ups to the in-process backup interception:
- Previously the matcher only allowed a bare `npx` prefix, so the
canonical zero-install form `npx -y runfusion.ai backup --create`
(and any `npx --yes` / `-p <pkg>` / `--package=<pkg>` variant) fell
through to the legacy shell-out path. The matcher now consumes any
number of npx flags before the binary token so all canonical
invocations route through the in-process executor.
- Previously the matcher accepted arbitrary text after `--create` and
the runner silently dropped it. Authors writing
`fn backup --create && notify-send done` or
`fn backup --create | tee log` reasonably expected the trailing
side effect to fire. The matcher now refuses any command containing
shell continuations / redirections / substitutions
(`&&`, `||`, `|`, `;`, `>`, `<`, backticks, `$()`), and rejects
trailing positional arguments. Such commands shell out as the user
wrote them.
The matcher is now a small tokenizer rather than a regex collection,
so the contract is easier to read and the unit-test grid covers each
permitted prefix combination plus all the previously-unhandled shell
forms.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Demote the refresh message from console.error to debugMcp so it no
longer appears as an error in normal output.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Defaults to true so existing behaviour (dashboard probes PATH for `fn` /
`fusion` and surfaces install / version-mismatch states) is unchanged.
When the user toggles it off in Settings → General → CLI Binary:
- `GET /system/fn-binary/status` short-circuits before `detectFnBinary()`
and returns `state: "skipped"`. No `<bin> --version` subprocess is
spawned at all, so an outdated globally-installed CLI cannot run as a
side effect of opening the dashboard.
- The install banner is hidden when `state` is `"skipped"`.
- `POST /system/fn-binary/install` rejects with HTTP 409 since install
is the user action that the status check informs.
- The Settings UI gains a checkbox under the existing CLI binary panel.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two follow-up corrections to the in-process auto-backup interception:
- The matcher previously hijacked any `fn backup …` / `fusion backup …` /
`runfusion.ai backup …` form. The in-process replacement only knows how
to do `--create` + cleanup, so scheduling `--list`, `--cleanup`, or
`--restore <file>` would have silently executed a create instead of the
requested operation. The matcher is now anchored to `backup --create`
(with optional trailing flags), with positive/negative unit tests.
- Step-based automations (`AutomationStep` with `type: "command"`) also
shell out — the legacy-command interception alone left that path
vulnerable. `executeCommandStep` now applies the same in-process backup
detour, factored through a shared `runBackupActionInProcess` helper.
Independently, `runProbe` in fn-binary now spawns with `cwd: tmpdir()`.
The dashboard's `/system/fn-binary/status` route runs `<bin> --version`
on whatever fusion binary happens to be on PATH — older releases (e.g.
v0.13.0) initialise an engine and create a fresh `.fusion/<project>/
.fusion/` tree as a side effect. Pinning the probe's cwd to the OS temp
directory keeps any such artefacts off the developer's project.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The backup automation was scheduled with `npx runfusion.ai backup --create`, which spawns whatever fusion binary is on PATH. On developer machines that's usually an older globally-installed runfusion.ai (v0.13.0 at time of writing) which still carries the pluginStore-rootDir bug — every backup tick recreated `<project>/.fusion/.fusion/` with a fresh empty TaskStore.
Cron-runner and routine-runner now intercept any command matching `fn backup`, `fusion backup`, or `npx runfusion.ai backup` and call `runBackupCommand` directly via the engine's open TaskStore. The interception also handles existing schedules persisted with the old npx command, so users do not need to manually update their automation rows.
The default command for newly created backup schedules is also simplified to `fn backup --create` — both forms route through the same in-process executor.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Throws if a caller passes a path whose last two segments are both `.fusion` — that pattern only happens when a Store class joins `.fusion` onto a path that already ends in `.fusion`. Surfaces the bug at the call site instead of silently creating a stray nested directory under the project. Complements the recent in-process-runtime PluginStore fix by catching any other call paths still passing the wrong rootDir.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
PluginStore expects a project-root path and appends `.fusion` itself, but the in-process runtime was handing it the already-resolved `.fusion` directory — producing a spurious `.fusion/.fusion/fusion.db` on every engine startup.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Merged feat(FN-3059) which aligns provider metadata and documentation across the codebase, updating README and getting-started docs plus refinements to the CustomProviderForm and ProviderIcon dashboard components.
Fusion-Task-Id: FN-3059
