[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1013":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":9,"rankLanguage":9,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":22,"topics":25,"createdAt":9,"pushedAt":9,"updatedAt":26,"readmeContent":27,"aiSummary":28,"trendingCount":15,"starSnapshotCount":15,"syncStatus":29,"lastSyncTime":30,"discoverSource":31},1013,"Fusion","Runfusion\u002FFusion","Runfusion","Multi node agent orchestrator - build faster and better",null,"TypeScript",701,81,3,40,0,20,46,233,60,9.74,"MIT License",false,"main",true,[],"2026-06-12 02:00:22","\u003Cdiv align=\"center\">\n\n\u003Cimg src=\".\u002Fdemo\u002Fassets\u002Ffusion-logo.png\" alt=\"Fusion\" width=\"120\" \u002F>\n\n# Fusion\n\n### From rough idea to production code — automatically.\n\n**Multi-node agent orchestrator** — tasks, agents, missions, git, files, and worktrees, with any model, local or cloud.\n\n[**runfusion.ai →**](https:\u002F\u002Frunfusion.ai) · [Docs](.\u002Fdocs\u002FREADME.md) · [GitHub](https:\u002F\u002Fgithub.com\u002FRunfusion\u002FFusion) · [npm](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@runfusion\u002Ffusion) · [Discord](https:\u002F\u002Fdiscord.gg\u002Fksrfuy7WYR)\n\n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-3fb950.svg)](.\u002FLICENSE)\n[![npm](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002F@runfusion\u002Ffusion.svg?color=58a6ff)](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@runfusion\u002Ffusion)\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-join-5865F2?logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002Fksrfuy7WYR)\n![Status](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fstatus-early%20preview-d29922.svg)\n![Shipping](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fshipping-weekly-bc8cff.svg)\n\n\u003Cbr \u002F>\n\n\u003Cimg src=\".\u002Fdemo\u002Fassets\u002Ffusion-reel.gif\" alt=\"Fusion reel: from rough idea to production code\" width=\"900\" \u002F>\n\n\u003Cbr \u002F>\n\u003Cbr \u002F>\n\n\u003Ca href=\"https:\u002F\u002Frunfusion.ai\">\n  \u003Cimg src=\"https:\u002F\u002Frunfusion.ai\u002Ffusion-dashboard.png\" alt=\"Fusion dashboard: Planning, Todo, In Progress, In Review, Done kanban columns with active task cards\" width=\"900\" \u002F>\n\u003C\u002Fa>\n\n\u003C\u002Fdiv>\n\n---\n\n## Your entire dev environment. On a single pane of glass.\n\nDescribe a task in plain language. A planning agent reads your project, understands context, and writes a full `PROMPT.md` plan — steps, file scope, acceptance criteria. Then Fusion plans, reviews, executes, and reviews again, in an isolated git worktree, with a human approval gate wherever you want one.\n\nOne board. Controlled from anywhere. Laptop, Mac mini, Linux server, cloud VM, phone — all connected.\n\n> Like Trello, but your tasks get specified, executed, and delivered by AI. Built on the great work of [dustinbyrne\u002Fkb](https:\u002F\u002Fgithub.com\u002Fdustinbyrne\u002Fkb).\n\n---\n\n## The flow\n\n```\n  ①  Describe          ②  Planning             ③  The board           ④  Isolated worktree\n  ─────────────        ─────────────         ─────────────          ─────────────────────\n  \"Add dark mode   →   Agent writes    →   Plan → Review →    →   fusion\u002FFN-123 branch\n   toggle to           PROMPT.md           Execute → Review        concurrent, zero\n   settings panel\"     (steps, scope,      (per step, until        file conflicts\n                       acceptance)         done)\n```\n\n### See every step, before the merge\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Frunfusion.ai\u002Fscreenshot-task-detail.png\" alt=\"Fusion task detail: workflow steps visible on an in-progress task with diffs and file changes\" width=\"820\" \u002F>\n\u003C\u002Fdiv>\n\nEvery task shows its plan, its reviews, its diffs, and its file changes in real time. Jump into an active task and nudge direction, tighten constraints, pause, or re-prompt.\n\n---\n\n## What makes it different\n\n|  |  |\n|---|---|\n| 🧠 **AI planning** | Describe a task in plain language. Planning agents turn it into a `PROMPT.md` plan with steps, file scope, and acceptance criteria. |\n| 🔁 **Workflow gates** | Plan → Review → Execute → Review on every step. Pre-merge gates block bad code; post-merge gates run informational checks. |\n| 🌳 **Worktree isolation** | Each task runs in its own branch and worktree (`fusion\u002F{task-id}`). Parallel tasks. Zero conflicts. |\n| ⚡ **Smart merge** | Passing every gate? Fusion squash-merges and moves on. Opt into manual approval anywhere. |\n| 🛰️ **Multi-node mesh** | Laptop, Mac mini, Linux server, cloud VM, phone — all synced. Desktop, mobile, web. |\n| 🧩 **Any model** | Anthropic, OpenAI, Ollama, and more. Local and cloud coexist. |\n| 🏢 **Agent companies** | Import pre-built teams — 440+ agents across 16 companies — and run them autonomously for weeks. |\n| 📬 **Inter-agent messaging** | Built-in mailbox between agents. Delegate, clarify, coordinate. |\n| 🗺️ **Missions** | Hierarchical planning (Mission → Milestone → Slice → Feature → Task) with autopilot and validation contracts. |\n| 🔬 **Research** | Bounded research runs with web search, GitHub, local docs, and LLM synthesis. Turn findings into tasks. ([Docs](.\u002Fdocs\u002Fresearch.md)) |\n| 🧪 **Self-improvement** | Agents reflect on their own output and update their prompts as they learn your codebase. |\n| 🔓 **Open source. MIT.** | No vendor lock-in. Run it on your own hardware. Shipping weekly. |\n\n---\n\n## How it works\n\n```mermaid\ngraph TD\n    H((You)) -->|rough idea| T[\"Planning\u003Cbr\u002F>\u003Ci>auto-planning\u003C\u002Fi>\"]\n    T --> TD[\"Todo\u003Cbr\u002F>\u003Ci>scheduled for execution\u003C\u002Fi>\"]\n    TD --> IP[\"In Progress\u003Cbr\u002F>\u003Ci>for each step:\u003Cbr\u002F>plan, review, execute, review\u003C\u002Fi>\"]\n\n    subgraph IP[\"In Progress\"]\n        direction TD\n        NS([Begin step]) --> P[Plan]\n        P --> R1{Review}\n        R1 -->|revise| P\n        R1 -->|approve| E[Execute]\n        E --> R2{Review}\n        R2 -->|revise| E\n        R2 -->|next step| NS\n        R2 -->|rethink| P\n    end\n\n    R2 -->|done| IR[\"In Review\u003Cbr\u002F>\u003Ci>ready to merge,\u003Cbr\u002F>or auto-complete\u003C\u002Fi>\"]\n    IR -->|direct squash merge\u003Cbr\u002F>or merged PR| D[\"Done\"]\n\n    style H fill:#161b22,stroke:#8b949e,color:#e6edf3\n    style T fill:#2d2006,stroke:#d29922,color:#d29922\n    style TD fill:#0d2044,stroke:#58a6ff,color:#58a6ff\n    style IP fill:#1a0d2e,stroke:#bc8cff,color:#bc8cff\n    style P fill:#1a0d2e,stroke:#bc8cff,color:#e6edf3\n    style R1 fill:#1a0d2e,stroke:#bc8cff,color:#e6edf3\n    style E fill:#1a0d2e,stroke:#bc8cff,color:#e6edf3\n    style R2 fill:#1a0d2e,stroke:#bc8cff,color:#e6edf3\n    style NS fill:#1a0d2e,stroke:#bc8cff,color:#bc8cff\n    style IR fill:#0d2d16,stroke:#3fb950,color:#3fb950\n    style D fill:#1a1a1a,stroke:#8b949e,color:#8b949e\n```\n\nTasks with dependencies are processed sequentially. Independent tasks run in parallel. Optionally require manual approval before tasks move from Planning to Todo (`requirePlanApproval` setting).\n\n---\n\n## Multi-node. One board. Every platform.\n\n\u003Cdiv align=\"center\">\n\n\u003Cimg src=\".\u002Fdemo\u002Fassets\u002Ffusion-mesh.gif\" alt=\"Fusion mesh: laptop, Mac mini, Linux server, cloud VM, phone — all synced\" width=\"820\" \u002F>\n\n\u003Cbr \u002F>\n\n![macOS](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FmacOS-000000?style=for-the-badge&logo=apple&logoColor=white)\n![Windows](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWindows-0078D6?style=for-the-badge&logo=windows&logoColor=white)\n![Linux](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLinux-FCC624?style=for-the-badge&logo=linux&logoColor=black)\n![Web](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb-4285F4?style=for-the-badge&logo=googlechrome&logoColor=white)\n![iOS](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FiOS-000000?style=for-the-badge&logo=apple&logoColor=white)\n![Android](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAndroid-3DDC84?style=for-the-badge&logo=android&logoColor=white)\n\n\u003C\u002Fdiv>\n\nLaptop, Mac mini, Linux server, cloud VM, phone — every node is a peer. Your task state, agents, logs, and diffs stay synchronized across the mesh. The same Fusion ships as:\n\n- 🖥️ **Desktop app** — Electron for **macOS** (Intel + Apple Silicon), **Windows** 10\u002F11, and **Linux**\n- 📱 **Mobile app** — Capacitor for **iOS\u002FiPadOS** and **Android** ([MOBILE.md](.\u002FMOBILE.md))\n- 🌐 **Web dashboard** — any modern browser, served from the `fn dashboard` daemon\n- 🔌 **CLI** — `fn` binary + extension for terminal-first workflows\n\nStart the daemon on any node, connect your other devices, and the board follows you everywhere.\n\n---\n\n## Run an agent company\n\n\u003Cdiv align=\"center\">\n\n\u003Cimg src=\".\u002Fdemo\u002Fassets\u002Ffusion-company-reel.gif\" alt=\"Fusion agent company: import a team, run it autonomously for weeks\" width=\"820\" \u002F>\n\n\u003C\u002Fdiv>\n\nImport a team. Run it autonomously for weeks. **440+ agents across 16 companies**, wired for missions, mailboxes, and inter-agent delegation.\n\n```bash\nnpx companies.sh add paperclipai\u002Fcompanies\u002Fgstack\n```\n\n---\n\n## Compatible with the tools you already use.\n\nFusion integrates with the tools you love. **Hermes**, **Paperclip**, and **OpenClaw** all ship as first-class plugins — route any workspace to whichever runtime fits the task. And any Paperclip agent-company imports with a single command.\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\".\u002Fdemo\u002Fassets\u002Fhermes-logo.svg\" alt=\"Hermes\" height=\"56\" \u002F>\n\u003C\u002Fdiv>\n\n### [Hermes](https:\u002F\u002Fhermes-agent.nousresearch.com) \u003Csub>`experimental`\u003C\u002Fsub>\n\n\u003Csub>Nous Research\u003C\u002Fsub>\n\nThe open-source autonomous agent from **Nous Research**. Install the Hermes plugin and run agents through Hermes for long-running, context-growing work — route any Fusion workspace to it.\n\n### OpenClaw \u003Csub>`experimental`\u003C\u002Fsub>\n\nOpenClaw runtime support is available as an experimental plugin (`fusion-plugin-openclaw-runtime`) for runtime discovery\u002Fconfiguration parity. Configure agents with `runtimeConfig.runtimeHint: \"openclaw\"` after installing the plugin.\n\n\u003Cbr \u002F>\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\".\u002Fdemo\u002Fassets\u002Fpaperclip-logo.svg\" alt=\"Paperclip\" height=\"56\" \u002F>\n\u003C\u002Fdiv>\n\n### [Paperclip](https:\u002F\u002Fpaperclip.ing) \u003Csub>`experimental`\u003C\u002Fsub>\n\n\u003Csub>paperclip.ing\u003C\u002Fsub>\n\nThe human control plane for AI labor. Install the Paperclip plugin to run agents through Paperclip inside Fusion.\n\nFusion also natively supports the **[`companies.sh`](https:\u002F\u002Fgithub.com\u002Fpaperclipai\u002Fcompanies)** agent-company standard: import a prebuilt team — **440+ agents across 16 companies** — and let them coordinate over Fusion's mailbox, missions, and workflow gates for weeks of autonomous work. Same company format, same agents, same skills as Paperclip.\n\n```bash\nnpx companies.sh add paperclipai\u002Fcompanies\u002Fgstack\n```\n\n\u003Cbr \u002F>\n\n> **Hermes**, **Paperclip**, and **OpenClaw** are **experimental** runtime plugins — APIs and wire formats may shift between minor releases.\n\n---\n\n## Quick start\n\n**Zero install, straight from npm:**\n\n```bash\nnpx runfusion.ai\n```\n\nThat launches the dashboard. Subcommands forward through: `npx runfusion.ai task create \"fix X\"`, `npx runfusion.ai --help`, etc. (Or verbosely: `npx @runfusion\u002Ffusion dashboard`.)\n\n**One-line installer** (macOS & Linux — auto-picks Homebrew, falls back to npm):\n\n```bash\ncurl -fsSL https:\u002F\u002Frunfusion.ai\u002Finstall.sh | sh\nfusion dashboard\n```\n\n**Homebrew** (macOS & Linux):\n\n```bash\nbrew tap runfusion\u002Ffusion\nbrew install fusion\nfusion dashboard            # or: fn dashboard\n```\n\nOr as a one-liner (auto-taps): `brew install runfusion\u002Ffusion\u002Ffusion`.\n\n**npm global**:\n\n```bash\nnpm install -g @runfusion\u002Ffusion\nfn dashboard                # or: fusion dashboard\n```\n\n**From a clone** (for development):\n\n```bash\npnpm dev dashboard\n```\n\nThen click the `Open:` URL printed in the terminal. It embeds a bearer token\n(`http:\u002F\u002Flocalhost:4040\u002F?token=fn_...`) that the browser captures to\n`localStorage` on first visit and reuses automatically thereafter. On the\nserver side, Fusion now persists the dashboard\u002Fdaemon token in\n`~\u002F.fusion\u002Fsettings.json` on first authenticated run and reuses it on later\nstarts unless you override it (`--token`, `FUSION_DASHBOARD_TOKEN`,\n`FUSION_DAEMON_TOKEN`) or disable auth with `--no-auth`. See\n[CLI reference → fn dashboard → Authentication](.\u002Fdocs\u002Fcli-reference.md#fn-dashboard)\nfor full precedence and reset\u002Frevocation options.\n\n### First-run setup\n\nOn first launch, Fusion opens the **onboarding wizard** with three guided steps:\n\n1. **AI Setup** — Use a simplified quick-start provider list (recommended providers plus any already-connected providers), then expand **Advanced provider settings** only if you need additional providers or setup details. You only need one provider to get started. Deprecated Google Gemini CLI \u002F Antigravity provider entries are intentionally hidden; Google\u002FGemini API key, Google Generative AI, Vertex, and Cloud Code paths remain supported.\n2. **GitHub (Optional)** — Connect GitHub for issue import and PR management\n3. **First Task** — Create your first task or import from GitHub (if no project is active, onboarding first prompts you to register\u002Fselect a project directory)\n\nThe wizard is **dismissible and non-blocking** — click **Skip for now** to use the dashboard immediately. Re-trigger it later from **Settings → Authentication → Reopen onboarding guide**.\n\n### Mobile\n\nFor Capacitor + PWA workflow, see [MOBILE.md](.\u002FMOBILE.md).\n\n---\n\n## Documentation\n\n| Guide | What it covers |\n|---|---|\n| [Getting Started](.\u002Fdocs\u002Fgetting-started.md) | Installation and onboarding |\n| [Dashboard Guide](.\u002Fdocs\u002Fdashboard-guide.md) | Board\u002Flist views, terminal, git manager |\n| [Task Management](.\u002Fdocs\u002Ftask-management.md) | Task lifecycle and CLI commands |\n| [Settings Reference](.\u002Fdocs\u002Fsettings-reference.md) | Configuration options |\n| [Architecture](.\u002Fdocs\u002Farchitecture.md) | System internals |\n| [Agents](.\u002Fdocs\u002Fagents.md) | Agent management, spawning, heartbeat |\n| [Workflow Steps](.\u002Fdocs\u002Fworkflow-steps.md) | Quality gates, templates, phases |\n| [Missions](.\u002Fdocs\u002Fmissions.md) | Mission hierarchy, planning, autopilot |\n| [Multi-Project](.\u002Fdocs\u002Fmulti-project.md) | Central registry, isolation modes |\n| [Docker](.\u002Fdocs\u002Fdocker.md) | Container deployment |\n\n---\n\n## Core features\n\n- **AI Planning** — Planning agent generates detailed `PROMPT.md` with steps, file scope, and acceptance criteria\n- **Step-by-step Execution** — Plan → Review → Execute → Review cycle for each task step\n- **Git Worktree Isolation** — Each task runs in its own worktree (`fusion\u002F{task-id}` branch)\n- **Workflow Steps** — Configurable quality gates (pre-merge: blocks merge; post-merge: informational)\n- **GitHub Integration** — Import issues, create PRs, real-time PR\u002Fissue badges\n- **Dashboard** — Real-time kanban board, agent management, terminal, git manager, mission planner\n- **Missions** — Hierarchical planning (Mission → Milestone → Slice → Feature → Task) with autopilot, validation contracts, fix-feature retries, and blocked-handoff semantics\n- **Multi-Project** — Manage multiple projects from a single installation with project isolation\n- **Inter-Agent Messaging** — Built-in messaging for coordination between agents and users\n\n### Provider authentication\n\nFusion supports OAuth-based authentication for AI providers configured via **Settings → Authentication**. For most OAuth providers, when the dashboard is accessed via a non-localhost host (remote node, LAN host\u002FIP, or reverse proxy), provider login URLs are rewritten to route OAuth callbacks through a bridge endpoint (`\u002Fapi\u002Fauth\u002Foauth-callback`) so redirects reach the active browser session.\n\n- **Anthropic (Claude)** — Uses a pasted authorization-code flow in Settings\u002Fonboarding: sign in, then paste the final redirect URL (or code) back into Fusion to complete login\n- **OpenAI Codex** — Uses the same pasted authorization-code flow with secure state validation\n- **Factory AI — via Droid CLI** *(optional)* — requires local Droid CLI install + `droid auth login`; detection follows the effective runtime binary path (default `droid`, or plugin `droidBinaryPath` when configured), then enable in **Settings → Authentication** and restart Fusion\n- **llama.cpp — via HTTP server** *(optional)* — configure your llama.cpp server URL (default `http:\u002F\u002F127.0.0.1:8080`) and optional API key, then enable in **Settings → Authentication**\n- **Other providers** — Authenticate via API key entry in Settings (including Google\u002FGemini API key, Google Generative AI, Vertex, and Cloud Code aliases)\n\n### Model system\n\nFusion uses a dual-scope model hierarchy with five independent lanes. Global settings define baseline defaults; project settings provide per-project overrides.\n\n| Lane | Purpose | Global Baseline Keys | Project Override Keys |\n|------|---------|---------------------|----------------------|\n| Executor | Task execution agent | `executionGlobalProvider` + `executionGlobalModelId` | `executionProvider` + `executionModelId` |\n| Planning | Task planning agent | `planningGlobalProvider` + `planningGlobalModelId` | `planningProvider` + `planningModelId` |\n| Validator | Plan\u002Fcode reviewer | `validatorGlobalProvider` + `validatorGlobalModelId` | `validatorProvider` + `validatorModelId` |\n| Title Summarization | Auto-title generation | `titleSummarizerGlobalProvider` + `titleSummarizerGlobalModelId` | `titleSummarizerProvider` + `titleSummarizerModelId` |\n| Workflow Step Refinement | AI prompt refinement | (uses `defaultProvider`\u002F`defaultModelId`) | (uses `modelProvider`\u002F`modelId` on WorkflowStep) |\n\n**Per-Task Overrides:** Tasks can override the executor, validator, and planning lanes with per-task model fields (`modelProvider`\u002F`modelId`, `validatorModelProvider`\u002F`validatorModelId`, `planningModelProvider`\u002F`planningModelId`).\n\n**Precedence:** Per-task → Project override → Global lane → `defaultProvider`\u002F`defaultModelId` → Automatic resolution.\n\nFor full settings documentation, see [Settings Reference](.\u002Fdocs\u002Fsettings-reference.md).\n\n### Scheduled tasks \u002F automations\n\nFusion supports scheduled task automation via the `\u002Fapi\u002Fautomations` endpoints. Automations can run shell commands or multi-step workflows on a configurable schedule.\n\n#### Scheduling scope\n\nAutomations and routines can run in two scopes:\n\n- **Global** — Runs across all projects. Use this for cross-project maintenance, backups, or unified reporting.\n- **Project** — Runs only within a specific project. Use this for project-specific CI, testing, or deployment tasks.\n\nWhen you create a schedule without choosing a scope, Fusion defaults to **project scope** with the `default` project ID for backward compatibility.\n\nTo explicitly target a scope:\n- In the dashboard **Scheduled Tasks** modal, use the **Global \u002F Project** toggle.\n- Via the API, pass `?scope=global` or `?scope=project&projectId=\u003Cid>` on automation\u002Froutine endpoints.\n\n**Scope resolution rules:**\n- `scope=global` always resolves to the global automation\u002Froutine lane, independent of the active project.\n- `scope=project` requires a `projectId`. If omitted, it falls back to `\"default\"`.\n- CRUD, run, toggle, and webhook operations are strictly scope-isolated: a global schedule cannot be mutated from a project-scoped request, and vice versa.\n\n**Operational guidance for multi-project setups:**\n- Prefer **global** schedules for shared infrastructure (e.g., nightly backups, memory insight extraction).\n- Prefer **project** schedules for per-repository automation (e.g., per-project test runners, deployment hooks).\n- Global and project lanes are polled independently by the engine, so due runs in one lane do not block the other.\n\n#### Automations\n\n| Endpoint | Method | Description |\n|---------|--------|-------------|\n| `\u002Fapi\u002Fautomations` | GET | List all automations (filtered by scope if specified) |\n| `\u002Fapi\u002Fautomations` | POST | Create automation (scope defaults to `project`) |\n| `\u002Fapi\u002Fautomations\u002F:id` | GET | Get automation by ID |\n| `\u002Fapi\u002Fautomations\u002F:id` | PATCH | Update automation |\n| `\u002Fapi\u002Fautomations\u002F:id` | DELETE | Delete automation |\n| `\u002Fapi\u002Fautomations\u002F:id\u002Frun` | POST | Trigger manual run |\n| `\u002Fapi\u002Fautomations\u002F:id\u002Ftoggle` | POST | Toggle enabled\u002Fdisabled |\n| `\u002Fapi\u002Fautomations\u002F:id\u002Fsteps\u002Freorder` | POST | Reorder automation steps |\n\n#### Routines\n\nRoutines are AI agent tasks triggered by cron schedules, webhooks, or manual execution. Routines share the same global\u002Fproject scope model as automations.\n\n| Endpoint | Method | Description |\n|---------|--------|-------------|\n| `\u002Fapi\u002Froutines` | GET | List all routines (filtered by scope if specified) |\n| `\u002Fapi\u002Froutines` | POST | Create routine (scope defaults to `project`) |\n| `\u002Fapi\u002Froutines\u002F:id` | GET | Get routine by ID |\n| `\u002Fapi\u002Froutines\u002F:id` | PATCH | Update routine |\n| `\u002Fapi\u002Froutines\u002F:id` | DELETE | Delete routine |\n| `\u002Fapi\u002Froutines\u002F:id\u002Frun` | POST | Manual trigger |\n| `\u002Fapi\u002Froutines\u002F:id\u002Ftrigger` | POST | Canonical manual trigger |\n| `\u002Fapi\u002Froutines\u002F:id\u002Fruns` | GET | Get execution history |\n| `\u002Fapi\u002Froutines\u002F:id\u002Fwebhook` | POST | Webhook trigger (signature verification supported) |\n\n---\n\n## CLI quick examples\n\n```bash\nfn task create \"Fix the login bug\"                    # Quick entry → planning\nfn task plan \"Build auth system\"                      # AI-guided planning\nfn task import owner\u002Frepo --labels bug                # Import GitHub issues\nfn task show FN-001                                   # View task details\nfn task logs FN-001 --follow                          # Stream execution logs\nfn task steer FN-001 \"Use TypeScript\"                 # Guide the agent mid-execution\n\nfn project add my-app \u002Fpath\u002Fto\u002Fapp                    # Register a project\nfn project list                                       # List all projects\n\nfn settings set maxConcurrent 4                       # Configure settings\nfn settings export                                    # Export configuration\n\nfn mission create \"Auth System\" \"Build auth\"          # Create mission\nfn mission activate-slice \u003Cslice-id>                  # Activate a slice\n\nfn skills search react                                # Search skills.sh\nfn skills install firebase\u002Fagent-skills               # Install agent skills\n```\n\n---\n\n## Packages\n\n| Package | Description |\n|---------|-------------|\n| `@fusion\u002Fcore` | Domain model — tasks, board columns, SQLite store |\n| `@fusion\u002Fdashboard` | Web UI — Express server + kanban board with SSE |\n| `@fusion\u002Fengine` | AI engine — planning, execution, scheduling, workflow steps |\n| `@runfusion\u002Ffusion` | CLI + extension — published to npm |\n\n---\n\n## Development\n\n```bash\npnpm install                  # Install dependencies\npnpm local                    # Start local dashboard\u002FAPI on a non-4040 port\npnpm local -- --engine        # Start local dashboard with the AI engine\npnpm build                    # Build default workspace packages (excludes desktop\u002Fmobile)\npnpm build:all                # Build all packages (including desktop\u002Fmobile)\npnpm dev dashboard            # Run dashboard + AI engine\npnpm dev:ui                   # Dashboard only (no AI engine)\npnpm lint                     # Lint all packages\npnpm typecheck                # Type-check all packages\npnpm test                     # Run all tests\n```\n\n### Build a standalone executable\n\nBuild a single self-contained `fn` binary using [Bun](https:\u002F\u002Fbun.sh\u002F):\n\n```bash\npnpm build:exe                # Build for current platform\npnpm build:exe:all            # Cross-compile for all platforms\n```\n\n---\n\n## License\n\nMIT — open source, no vendor lock-in. See [LICENSE](.\u002FLICENSE).\n\n\u003Cdiv align=\"center\">\n\n**[runfusion.ai →](https:\u002F\u002Frunfusion.ai)**\n\n\u003C\u002Fdiv>\n","Fusion 是一个多节点代理编排工具，旨在加速和优化开发流程。它支持任务描述、自动规划、执行与审查等核心功能，并能够无缝集成 Git、文件管理和工作树等功能，适用于本地或云端模型。该工具使用 TypeScript 编写，具有高度可扩展性和灵活性。Fusion 适合需要从初步想法快速过渡到生产代码的开发团队使用，尤其在多环境（如笔记本电脑、服务器、云虚拟机）下协作时表现优异。通过提供一个统一的控制面板，Fusion 能够简化复杂的开发流程，提高工作效率。",2,"2026-06-11 02:41:07","CREATED_QUERY"]