This merge adds a new llama.cpp local AI provider integration, including a `pi-llama-cpp` CLI extension package with model resolution and context retrieval, a `LlamaCppProviderCard` UI component for provider configuration, backend API routes for model registration and llama.cpp probe/health checking
Fusion-Task-Id: FN-3377
Fixes FN-3286 by removing a duplicate PR creation success toast notification from the task detail modal. The change is a single line deletion in `TaskDetailModal.tsx`.
Fusion-Task-Id: FN-3286
Icons in the installed PWA were overlapped by the iOS home indicator and a
body-colored gap appeared between the nav bar and the executor status bar
(whose `bottom` already includes safe-area-inset-bottom + standalone-bottom-gap).
Adding matching `padding-bottom` to `.mobile-nav-bar` lifts icons above the
indicator and closes the gap.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
useTasks.deleteTask had a single-arg signature and dropped the options
object, so the modal's "remove dependency references" confirmation
flag never reached the API and deletion still failed with
TASK_HAS_DEPENDENTS. Forward options through the hook and widen the
matching taskOperations.deleteTask type in AppModals.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Moonshot/Kimi returns a non-standard finish_reason of "repeat" when its
server-side repetition detector trips. pi-ai surfaces this as a fatal
session state error, killing the agent heartbeat. Filter that case in
promptSessionAndCheck so the truncated turn is logged and the heartbeat
continues on the next tick.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The merge adds Step 2 of the native externalization hardening for the CLI package, including a minor changeset for `@runfusion/fusion` and tests validating bundle output and package configuration behavior. The changes strengthen the contract governing how native modules are handled during the build
Fusion-Task-Id: FN-3262
Fixes FN-3112: Updates NodesView CSS to use tokenized border-radius and transition values from the design system instead of hardcoded pixel values, ensuring consistency with the dashboard's established token variables.
Fusion-Task-Id: FN-3112
Merges FN-3237 to stabilize the mobile viewport teardown flow when closing the planning modal, ensuring the viewport state resets correctly. Adds test coverage for the mobile planning close path and includes a changeset for the patch release.
Fusion-Task-Id: FN-3237
Fixes task detail modal danger/error styling by switching to token-based CSS variables for consistent theming across dark and light modes.
Fusion-Task-Id: FN-3350
This merge adds a changeset for feature FN-3349 (in-review retry behavior) and delivery documentation to the `@runfusion/fusion` package.
Fusion-Task-Id: FN-3349
useMobileKeyboard now requires a focused input for keyboardOpen=true (both
the chrome-overlap and iOS-gap paths). The moment an input blurs, the hook
reports keyboardOpen=false instead of waiting hundreds of ms for iOS's
visualViewport dismissal animation to settle.
This makes App-level mobileKeyboardOpen flip false instantly on blur, so
MobileNavBar reappears and project-content regains nav-bar padding in the
same frame. The ChatView composer (and TodoModal/PlanningModeModal) snap
to their post-keyboard layout in one move instead of crawling down with
iOS's keyboard slide.
Replaces the per-component 450ms suppress hack in ChatView (also dropped
in this commit) which couldn't reach the parent layout's nav padding
state and produced "below tab bar then snap up" jitter.
Adds a regression test and updates four existing tests that assumed
"vv shrinks → keyboard up" without focus (focus is now required).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Add useMobileScrollLock hook using position:fixed body lock (Bootstrap/
Headless UI/Stripe pattern) to prevent iOS Safari from shifting the
document and visualViewport when an input inside a fixed-position modal
is focused. Wire into 15 input-bearing modals plus ChatView, replacing
ChatView's inline body-overflow effect.
- Widen computeBuildVersion in vite.config.ts to hash the entire app/ tree
so the version-check poll actually notices rebuilds (FN-3333 follow-up;
previously only main.tsx and package.json were hashed).
- ChatView: on input blur, suppress keyboard-aware sizing for 450ms while
reserving mobile-nav-bar space, so the composer snaps to its final
height in one move instead of crawling down with iOS's keyboard slide
and then jumping again when the nav bar reappears.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Heartbeat read non-existent split modelProvider/modelId fields while the
dashboard saves runtimeConfig.model as combined "provider/modelId", so
sessions fell through to pi's default model and failed with
"No API key for provider: openai-codex". Add extractRuntimeModel helper
that prefers the combined string and use it from heartbeat.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Merges FN-3341 mobile keyboard handling improvements into the dashboard, wiring `useMobileKeyboard` into both TodoModal and PlanningModeModal with keyboard-aware CSS overrides for mobile, plus regression tests for both components. FN-3337's chat `isGenerating` state recovery is also included.
Fusion-Task-Id: FN-3341
A task whose previous run exhausted its merge budget (mergeRetries=MAX)
could land back in in-review with status=null, where the merger refused
it (canMergeTask false) and the ghost-review fallback bounced it back to
todo every taskStuckTimeoutMs (10 min) — beating the 30 min merge
cooldown reset. Each fresh execution now starts with mergeRetries=0.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge includes three features: FN-3337 fixes mobile keyboard handling in the PlanningModeModal by wiring in `useMobileKeyboard` and adding keyboard-aware CSS sizing, with regression tests; FN-3336 recovers the `isGenerating` chat state on session reload to prevent UI flickering; and FN-3338 fix
Fusion-Task-Id: FN-3337
Merges FN-3336 to recover chat streaming state on page reload — adds `isGenerating()` and `getGeneratingSessionIds()` to `ChatManager`, enriches the session API to surface active-streaming sessions, and makes `useChat` and `useQuickChat` hooks restore the `isGenerating` flag when loading existing se
Fusion-Task-Id: FN-3336
The merge adds an auto-reload setting (FN-3334) with UI controls in the settings modal, documentation, and a new version-check module, while also fixing a bug (FN-3338) where extension providers incorrectly resolved the project root when invoked from git worktrees — moving the project-root resolutio
Fusion-Task-Id: FN-3338
Merged two features: FN-3334 introduces a new `autoReloadOnVersionChange` setting with a toggle in the Settings modal, version-check logic in the dashboard, and corresponding tests; FN-3335 fixes worktree project root resolution in `createFnAgent` and the skill resolver, with tests for both paths. S
Fusion-Task-Id: FN-3334
Merges FN-3335 (worktree project resolution) and FN-3333 (spurious version reloads). The engine's `createFnAgent` now resolves project root from the worktree's cwd rather than the parent process, with `resolveProjectRoot` added to skill-resolver for consistency. The dashboard's `versionCheck` was up
Fusion-Task-Id: FN-3335
Fixes spurious new-version reloads by making the build version deterministic using the git hash and adding a 60-second cooldown to the version check. Includes unit tests for the version check logic.
Fusion-Task-Id: FN-3333
The merge lands the mesh configuration system (FN-3111) with type definitions, a `MeshConfigGenerator` service using a new `DockerClientService.recreateContainer()` method, and backend API routes for managing Docker-based mesh nodes. It also adds a "New Chat" button to the ChatView header on desktop
Fusion-Task-Id: FN-3111
The verification-fix agent prompt previously forbade modifying files
unrelated to the failure, which blocked the natural fix when
deterministic merge verification failed because of stale/missing
plugin `dist/` outputs in sibling workspace packages (e.g.
`Failed to resolve import "./cli-spawn.js"` from
`fusion-plugin-hermes-runtime/dist`).
- Add explicit guidance to detect stale-artifact failure signatures
and rebuild the affected package(s) before editing source.
- Allow the agent to fix pre-existing breakage on the base branch,
preferring the smallest change that makes verification green.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge introduces visibility of the AI's first-turn reasoning in the planning modal (FN-3274), with corresponding regression tests, and removes all PI (Prompt Intelligence) install references and documentation from the codebase (FN-3309).
Fusion-Task-Id: FN-3274
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 (c9043519) 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>
