pi-ai's openai-codex-responses provider surfaces ChatGPT-plan WebSocket
drops as bare "WebSocket error" / "WebSocket closed <code>". The
underlying ErrorEvent.error is dropped by extractWebSocketError (it only
inspects event.message), so the cause is gone by the time we see it.
The engine then propagated these as fatal failures instead of retrying
a transient network blip.
- Add WebSocket patterns to TRANSIENT_ERROR_PATTERNS so the executor's
existing transient retry path picks them up.
- Tag the model id onto the thrown error and emit a structured warn in
promptSessionAndCheck so future drops can be triaged by transport.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge enables deletion of paused agents across the dashboard UI — adding delete controls to the AgentDetailView, AgentListModal, and AgentsView, with tests covering all three flows and a final typecheck compatibility fix.
Fusion-Task-Id: FN-3322
This merge lands four major features: the desktop app gains shell onboarding with remote mode support via a new `DesktopModeChooser` and `shell-settings` module (FN-3399); the dashboard gains full archived insights support with the `InsightsView` redesign and `useInsights` hook overhaul (FN-3315); C
Fusion-Task-Id: FN-3434
Adds test coverage for archived insights across the core store, `InsightsView` component, and `useInsights` hook — 156 lines of new tests covering FN-3315.
Fusion-Task-Id: FN-3315
Restores build stability in the dashboard workspace by updating `vite.config.ts` and `Board.tsx`, while updating the routes-auth test fixtures from anthropic to github-copilot models.
Fusion-Task-Id: FN-3417
Removes the standalone `DesktopShellBootstrap` component from the desktop package, routing desktop remote mode into the shell onboarding flow instead. Updates the corresponding test file to reflect the component removal and adjusts the README.
Fusion-Task-Id: FN-3399
Updates authentication route test fixtures to use "github-copilot" as the provider name instead of the previous value, ensuring tests align with the current auth provider configuration.
Fusion-Task-Id: FN-3413
Completes the commit body rendering in expanded git views within `GitManagerModal`, including updated component implementation and tests. The dashboard git routes and modal component now properly display full commit body content.
Fusion-Task-Id: FN-3301
Merges shell connection infrastructure across desktop and mobile (NativeShellConnectionManager, NativeShellOnboardingModal, connection-profiles and native-shell plugins), strengthens the research store lifecycle with expanded test coverage, enhances WorkflowStepManager with new UI, and adds document
Fusion-Task-Id: FN-3457
The merge lands Step 7 of FN-3398, adding documentation and delivery artifacts across the mobile and desktop packages with updated READMEs, mobile-specific docs, and architecture references.
Fusion-Task-Id: FN-3398
Docs(FN-3397): Documented the add-step chooser flow in the workflow steps reference, adding 9 lines to `docs/workflow-steps.md`.
Fusion-Task-Id: FN-3397
The merge lands FN-3370's research store hardening work, updating the preflight documentation with a refined coverage scope and adding regression tests for the research store lifecycle alongside the corresponding implementation adjustments in `packages/core/src/research-store.ts`.
Fusion-Task-Id: FN-3370
The project's ESLint flat config does not include the react-hooks plugin,
so referencing its rules in eslint-disable comments causes a "rule not found"
error. Replace with a plain comment preserving the intent.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Split the combined useEffect that handled both auto-start and
persisted-goal restoration into two separate effects:
- Goal restoration: depends only on [isOpen], runs once on modal open
- Auto-start: keeps handleStartInterview dep for initialGoalProp flow
Previously the single effect depended on handleStartInterview, which
recreates on every missionGoal change. Every keystroke re-triggered
the effect, which read the stale persisted value from localStorage
and overwrote user input — making the textarea uneditable and showing
stale data from the last planned mission.
Co-authored-by: AI <ai@runfusion.ai>
Co-authored-by: Orca <help@stably.ai>
This merge delivers three distinct features: a compact mobile header row for the dashboard with corresponding regression tests, a plugin slot registry with Docker node provisioning for the mesh network, and heartbeat agent sessions now support tool mode for expanded capability access.
Fusion-Task-Id: FN-3414
This merge introduces a plugin slot registry system for extensible dashboard panels, adds docker node provisioning routes with planning modals, enables tool mode for heartbeat agent sessions, and enhances the agent detail view with skill badges. It also includes research settings documentation, sett
Fusion-Task-Id: FN-3115
This merge enables heartbeat agents to use coding sessions with tool access (FN-3459). It adds heartbeat session prompt tests, updates the heartbeat executor to support tool mode, modifies dashboard node components and hooks, and includes a fix for a typecheck regression. A changeset is included for
Fusion-Task-Id: FN-3459
Adds readable skill badges to the agent detail view with an interactive detail panel, along with corresponding tests and utility functions. A typecheck fix for the dashboard docker config is included.
Fusion-Task-Id: FN-3381
This merge lands five features spanning the Fusion stack: research settings key renaming (FN-3313, Steps 2–7) across types, defaults, CLI, and engine packages; static plugin slot-host rendering contract documentation (FN-3260); Docker node provisioning routes with a planning modal (FN-3116); plannin
Fusion-Task-Id: FN-3313
Documents the static slot-host rendering contract in `docs/PLUGIN_AUTHORING.md`, clarifying how plugins should interact with the rendering system.
Fusion-Task-Id: FN-3260
This merge adds Docker node provisioning infrastructure with API routes for provisioning lifecycle, Docker node management, and failure-state handling, along with corresponding test coverage for onboarding, lifecycle, and error flows. It also introduces planning session rewind capability with a moda
Fusion-Task-Id: FN-3116
This merge delivers four major features: a planning session rewind system (FN-3447, steps 1–4) with a new backend route for rolling back sessions, modal back-action wiring, and updated typing; workspace verification gates (FN-3385) for agent prompt editing; an agents view org chart spacing rework (F
Fusion-Task-Id: FN-3447
- store.ts close(): clear retry timer and buffer before nulling _db to
prevent re-opening a new connection after close (Greptile P1)
- merger-staging-allowlist.test.ts: use `git init -b main` so the
default branch is always named 'main' regardless of git config
- SettingsModal.test.tsx: fix expected text for compact DroidCli card
— shows "✓ Active" badge, not "✓ Connected — 1.2.3"
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Reintroduce local task-priority sorting helpers in Board to avoid cross-package runtime dependency coupling
- Keep board ordering deterministic by sorting by priority, creation time, then numeric task ID
- Document rich prompt editing affordances for Soul, Agent Memory, and Inline Instructions in agent creation flow
Fusion-Task-Id: FN-3385
Rebuilt the agents/org chart layout with improved spacing, adding corresponding tests for the updated card styling and mobile responsiveness.
Fusion-Task-Id: FN-3375
This merge delivers three focused changes: fixes the Git Manager modal mobile overflow (FN-3367) with new CSS, stabilizes the Docker config modal recreate indicator tests (FN-3114) with comprehensive new test coverage, and adds hook baseline cleanup tests (FN-3362). The changes span the core databas
Fusion-Task-Id: FN-3362
The merge stabilizes a test case in the NodeDetailModal test file related to the docker config modal's recreate indicator, correcting an assertion that was causing intermittent failures.
Fusion-Task-Id: FN-3114
- Add mobile-only Git Manager layout rules so changes lists and sections can shrink without horizontal page scrolling
- Wrap section header actions and file rows at narrow widths while preserving readable file names and controls
- Extend GitManagerModal CSS regression coverage for mobile wrapping selectors
- Add core mobile modal test assertions for wrapped actions/rows and bounded section width
- Add a patch changeset for @runfusion/fusion documenting the mobile overflow fix
Fusion-Task-Id: FN-3367
Stabilizes mobile keyboard tests in ChatView by refining timing assertions and test setup, addressing FN-3357 regressions in the third and final step of the fix.
Fusion-Task-Id: FN-3357
Refactored ListView component CSS to use design token variables instead of hardcoded values, ensuring consistent styling with the dashboard's design system across all themes.
Fusion-Task-Id: FN-3445
Default mouse reporting OFF so click-drag selection works by default —
useful for copying the auth token directly off the System panel without
needing the [c] keyboard shortcut. Auto-enable when the user focuses a
panel that consumes wheel events:
- Status mode → on while Logs is focused, off elsewhere
- Interactive → on for Files / Git / Board (which uses wheel in the
task-detail screen), off for Agents / Settings
Implemented as a useEffect in DashboardApp that maps state.activeSection
+ state.mode + state.interactiveView → controller.setMouseEnabled. [M]
remains a manual override; the next focus change reapplies the auto
policy. Controller's start() now honors initial mouseEnabled rather than
unconditionally writing the SGR enable sequence.
Updated System-panel hint to advertise click-drag selection, simplified
the help-overlay [M] entry, and flipped the createInitialState test
expectation.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge delivers v0.19.0 with several major features and reliability improvements: a new llama.cpp provider integration for the pi extension and mobile dashboards, an extensive refactor of the agent heartbeat system with per-agent configuration and test coverage, a large PlanningModeModal test su
Fusion-Task-Id: FN-3356
Adds comprehensive tests for cleanup reliability in the in-process runtime (131-line test suite) plus related coverage in the executor tests, with a small fix to the runtime implementation itself to address the reliability issue.
Fusion-Task-Id: FN-3352
Adds `wait_for_completion` polling support for research runs in the CLI extension, along with tests covering the research status schema and wait behavior, and ships a patch changeset for `@runfusion/fusion`.
Fusion-Task-Id: FN-3330
The merger issues several `git reset --hard` / `git reset --merge` and
forced-checkout calls against `rootDir` during merge attempts. When
`rootDir` is the developer's primary checkout (common for solo / single-host
setups), those resets silently discard any unrelated unstaged or untracked
changes — we burned dev work this way during FN-3329 (dashboard-tui edits
wiped mid-flight by an unrelated merge run).
Snapshot dirty paths at entry to `aiMergeTask`, stash them under a
recognizable label including the taskId (`-u` to capture untracked), and
pop them in a finally block on every exit path. On pop conflict we leave
the stash intact and log a recovery hint rather than dropping it. Best-
effort: a stash failure logs and proceeds with the old behavior so the
merge itself is never blocked.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Expand dashboard TUI state/controller wiring for system panel interactions and mouse toggle behavior
- Update dashboard TUI app rendering and tests to cover the new panel controls
- Improve SetupWizardModal layout and interaction behavior with matching CSS and test updates
- Restore docs/research.md wording and add a changeset for the published CLI package
Fusion-Task-Id: FN-3329
Merged FN-3328: Refactored the droid integration by converting `droid-cli` into a lightweight compatibility shim that delegates to `fusion-plugin-droid-runtime`, reducing droid-cli by ~5,400 lines of code while moving the actual runtime logic into the plugin scaffold. Updated the plugin loader to al
Fusion-Task-Id: FN-3328
Removed the monolithic PlanningModeModal test file (3,476 lines), completing the refactor of the test suite for this component.
Fusion-Task-Id: FN-3325
The merge lands FN-3323's self-healing system: a new "reports health" prompt section and `agent-heartbeat.ts` module that detect orphaned agents and recover their state (clearing stale error states and re-enabling heartbeat execution). The engine's `self-healing.ts` was expanded with wired recovery
Fusion-Task-Id: FN-3323
- Add shared startup model sync helper and integrate it into dashboard, serve, and daemon command startup flows
- Introduce opencode-go model sync setting in core schema/types and expose it in dashboard settings UI
- Add CLI regression tests covering startup model sync behavior and command wiring
- Document the new startup model sync setting and related CLI behavior updates
Fusion-Task-Id: FN-3227
