Runfusion/Fusion v0.58.0

on GitHub

New

• Add an interactive worktree-rooted Terminal tab to the task detail view.
• The Task Detail Terminal tab is now always available, falling back to the project root when a task has no worktree.
• Task Detail terminal now shows its worktree, is shorter on mobile, and sits with Cost after Comments.
• Agents can now add files to a task's File Scope while working, so out-of-scope edits aren't stranded at merge.
• Agents save screenshots, videos, HTML mockups, and PDFs as artifacts, shown in a new category gallery with doc editing.
• Edit task documents and project files in the Artifacts view; markdown by default; fix the Add comment button.
• Add a persistent Advanced settings toggle that keeps uncommon Settings sections and controls hidden by default.
• Guide repeatable Compound Engineering cycles from product grounding through reusable learnings.
• Add persisted thinking-level settings for every fallback model lane.
• Cursor CLI model-picker rows now surface reasoning/context-window metadata when the Cursor CLI reports it.
• Add Grok CLI runtime support as a bundled plugin with a grok-cli model provider.
• Mobile Settings search row now collapses by default with a show/hide toggle.
• Add a policy-gated review-lane bypass for cards stranded by a failed pre-merge review step.
• Grok can now run through the Grok CLI's NDJSON stream, so CLI-authenticated setups need no Fusion-visible API key.
• Grok CLI runtime now bridges tool execution events (name/args/result) from the NDJSON stream, not just text.
• Grok work can now be routed through the Grok CLI streaming runtime, not only the direct xAI endpoint.
• Add a dedicated permission for who may bypass a failed review gate, separate from task mutations.
• Add a File Scope agent permission category, allowed by default under the grant-all preset.
• Update the pi SDK and add support for GPT-5.6 codex-tier models.
• GPT-5.6 codenamed models (luna, sol, terra) are now selectable in the model picker.
• Add inline thinking-level selection to task and agent model dropdowns.
• Add per-lane thinking effort overrides to Settings model lane dropdowns.
• Add per-node workflow thinking-level controls for custom model bindings.
• Add per-workflow model lane thinking-level controls for planning, execution, and review.
• Choose a thinking level when starting a new model chat.
• Plugin prompt contributions can now gate content on per-project plugin settings.
• Let task edits toggle optional workflow steps directly.
• Show a Claude "Weekly (Fable)" usage window in the Usage dropdown.
• Image attachments now appear in the Artifacts view as artifacts.
• Apply fallback models' own thinking levels when runtime swaps to them.
• Add inline thinking-level selectors to every fallback model picker in Settings.
• Quick-add composer now shows icon-only priority/Fast controls with GitHub tracking beside attach.
• Artifacts view — Task Documents now uses a left-sidebar list with a right-pane content viewer.
• Artifacts view — select text in a Task Document's content pane to comment and send it to a new task.
• Show a Grok (xAI) card in the Usage dropdown for configured Grok API keys.
• Show Cursor subscription usage in the Usage dropdown.
• Add a documented CURSOR_API_KEY credential path for Cursor usage metering.
• Add a Cost tab to task detail and an optional per-card cost badge (default off).
• Add a resizable Settings navigation rail that remembers its width.
• Move terminal shortcuts into the footer and collapse crowded terminal tabs into a dropdown.
• Show task Artifacts-tab documents expanded with Markdown by default.
• CLI agent cold-start timeouts now default to 2 minutes and are configurable.
• Done task cards group Archive and Revert into one dropdown.
• Priority selection in quick add and task cards is now color-coded by urgency (blue low, amber high, red urgent).
• Artifacts view — the Task Documents list now also shows each task's registered artifacts.
• Add cost data for GLM-5.2, MiniMax-M3, and Kimi K2.6 so their token usage shows a dollar cost instead of "—".
• Drive Grok CLI sessions over ACP with Fusion tools, skills, and MCP loaded.
• Add a simplified workflow editor view with a modern vertical canvas, plus Simple/Advanced/List mode toggle.
• Task cards now show a "Reviewing" badge while a task is in Plan Review.
• Usage dropdown now shows the Claude Fable weekly window and Grok CLI subscription credit usage.

Fixed

• Terminal now auto-reconnects on first launch instead of getting stuck on "Disconnected".
• Task card size badge (S/M/L) no longer drops onto a misaligned second row on cards with extra badges.
• Task card size badge (S/M/L) now sits flush against the card's right edge.
• Stop false "Anthropic OAuth expired" notifications when the token is actually valid.
• Fix the Artifacts preview "Add comment" button doing nothing when clicked, and label the preview as read-only.
• Tidy the board quick-add composer and add a visible task-card actions menu.
• Show partial estimated costs across all Command Center cost views when some model pricing is unavailable.
• Daemon exits non-zero on signal termination so Restart=on-failure restarts it after a memory-pressure kill.
• Prevent stale Planning notifications from pointing to missing sessions.
• Fix the GitHub/GitLab import preview panel being cut off on tablet-width screens.
• Fix misaligned padding on the Cursor CLI authentication card.
• Cursor CLI models now appear in Fusion's model picker when the Cursor CLI provider is enabled.
• Fix Cursor CLI model discovery and auth to use the real cursor-agent commands.
• The Cursor CLI binary path override now also applies to the model picker, not just sign-in/status.
• Fix the search icon overlapping typed/placeholder text in the Files — Project search input.
• Fix fn agent stop/fn agent start hanging up to 60s per retry instead of exiting.
• Background memory-index refresh no longer keeps short-lived CLI/Node processes alive.
• qmd-backed project memory search no longer keeps short-lived CLI/Node processes alive.
• Fix background SQLite integrity checks holding short-lived CLI commands open unnecessarily.
• Grok and Cursor CLI models now appear in model pickers immediately after enabling the provider.
• Grok CLI models now run instead of failing with "not found in the pi model registry".
• Fix Grok CLI model picker showing prompt text instead of real model names.
• Grok now uses the key from ~/.grok/user-settings.json when GROK_API_KEY is not set.
• Grok CLI no longer requires a Fusion-visible API key — the CLI's own auth is enough to enable it.
• Archiving a task now releases its active-session lock so the next task can run Plan Review.
• Fix agents needing repeated stop/start because a stopped agent's heartbeat timer was never fully cleared.
• Triage recovers automatically when the planning model hits a provider 404/429 and no fallback is set.
• Bound durable-agent heartbeat worktree-acquisition retries and count exhausted failures.
• The engine now reacts to CLI fn agent stop/start promptly instead of waiting up to a minute for the audit sweep.
• Preserve prior failed review-step attempts so self-healing re-runs no longer erase the failure history.
• Board mutations from a tool session no longer silently land in the wrong project database.
• fn task show/move now retry through a momentarily locked board database instead of failing.
• All fn task subcommands now retry a momentarily locked board database and exit promptly instead of hanging or leaking.
• Recovery and oversight now wait for approval-blocked tasks instead of resuming them early.
• fn branch-group/fn pr now retry a locked board database and exit promptly instead of hanging or leaking.
• fn backup/memory-backup/mcp/db vacuum now retry a locked board database and exit promptly instead of hanging.
• CLI research/settings-import/agent-export/git/project commands close board stores promptly and retry a locked database.
• The planner overseer now detects and recovers stalled in-progress tasks instead of leaving hung executors stuck.
• Harden the dashboard server so provider API keys keep persisting even if a host forgets to wire auth storage.
• Fix manual merge hold tasks being marked failed when auto-merge is off.
• Manual merge hold now applies to shared-branch-group tasks whose group has dissolved.
• Mobile Settings search icon now sits inline next to the section dropdown.
• Tighten the mobile Settings layout — dropdown-only section picker, single-row footer, and slimmer header/footer.
• Grok CLI models now run via the grok CLI when no Fusion-visible API key is set.
• The latest OpenAI GPT-5.6 models now appear everywhere, not just the Settings model list.
• Fix empty estimated cost on the dashboard so priced runs show a dollar amount.
• Route no-key Grok CLI chat and fallback model selections through the bundled CLI runtime.
• GPT-5.6 codex models (luna, sol, terra) now actually appear in the codex model picker.
• Fix mobile model drop-down lists so they scroll by touch.
• Route Grok CLI models through the logged-in grok CLI in packaged hosts without requiring GROK_API_KEY.
• Fix oversized icons and spacing on the MCP servers settings page.
• Fix Grok CLI chat failing instantly with a "Response failed" error.
• Fix the Artifacts tab count for default-scope dashboards.
• Fix mobile Settings footer spacing so version text no longer overlaps actions.
• Style the Thinking Level dropdown to match the dark model picker across all surfaces.
• Fix Skills view showing "Skill not found" when opening any skill's content.
• Plugin skills now show for the project that enabled them, even when the daemon starts elsewhere.
• Show a No message placeholder for empty assistant chat replies.
• Grok CLI failures now show the actual error instead of an empty chat message.
• Task cards no longer wrap the header when a task was created by an agent — the agent badge moved to a bottom row.
• Surface Grok CLI runtime failures instead of empty chat replies.
• Fix Chat header showing thread controls while the conversation list is displayed after re-entering Chat.
• Tighten margins and padding across all mobile Settings pages for a more compact layout.
• Estimated cost on the dashboard stays populated as runtime model catalogs drift.
• Surface Grok CLI immediate no-message exits with actionable diagnostics.
• Fix the Thinking Level dropdown showing an unstyled white control in model pickers.
• Fix Grok CLI runtime sends to stream responses from xAI's real grok binary.
• Make Grok CLI chat replies reliable by using the stable headless JSON response.
• Usage view now shows meters only for AI providers you have configured.
• Fix a false "Project directory is not a Git repository" error that blocked all task execution in valid repos.
• Show the xAI logo for Grok model IDs across dashboard provider surfaces.
• Recover tasks stranded by missing worktrees during merge/review and allow retry.
• Mobile Settings footer shows the compact "v0.x" version instead of the full word.
• Usage view now hides Gemini when it isn't configured for metering or its login has expired.
• Fix the List view controls and quick-add box being cut off on tablet-width screens.
• Use the real Cursor logo in the usage dropdown, model selection, and other provider surfaces.
• Stop false "OAuth token expired" push notifications for providers that silently refresh (e.g. GitHub Copilot).
• Fix terminal header wrapping and spacing when the panel is narrow.
• Artifacts view — Task Documents sidebar now shows clearer task grouping and more space between tasks.
• Agents now auto-clear error state and retry on their next heartbeat instead of getting stuck.
• Fix the Artifacts Task Documents list rendering as blank rows when documents are loaded.
• Stop recording advisory "merger awaiting-confirmation" planner interventions that never block auto-merge.
• Fix the mobile Todo view so the list panel fills full height on selection.
• Coordinate durable-agent error recovery across heartbeat and self-healing.
• Fix push to remote after merge never running; pick the push remote and target branch from dropdowns in settings.
• Make stranded AI merge recovery bind to the reviewed clean-room commit.
• Honor assigned agent models in execution and warn on default-model fallbacks.
• Fix cramped padding on the Notifications "Failure notification mode" settings card.
• Fix Grok CLI chat returning errors or empty replies in the dashboard.
• Fix header connection pill showing "Desktop Desktop" and mixed font sizes.
• Fix Memory insights parsing and modernize the Memory, Insights, Todos, and agent Memory views.
• Polish the Memory view: centered layout, labeled editor toolbar, aligned toggle rows.
• Reconcile completed and stale generated-fix mission invariants.
• Settings on mobile now keeps showing the GitHub star count.
• Keep task detail per-model cost tables horizontally scrollable on mobile.
• Tasks are no longer stuck "awaiting release authorization" — the over-firing release gate was removed.
• Fix Settings GitLab row overflowing its panel and the footer Save button clipping.
• Polish first-run setup: connected providers first, state-driven GitHub step, fixed radios, deduped node picker.
• Auto-heal wedged SQLite connections in place instead of failing every request until restart.
• Task API operations no longer fail with 500 when a task's PROMPT.md can't be read; server also logs 500 causes.
• "Update now" now explains permission (EACCES) failures and how to fix them instead of showing raw npm errors.

Performance

• Fix a slow dashboard memory leak where archived tasks were never evicted from the in-memory badge cache.

Internal

• Removed leftover UI/i18n/docs for the deleted "awaiting release authorization" planning hold.
• Update the bundled pi SDK to 0.80.6.
• Document the planner-overseer eye badge on task cards.