The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Add Claude DevFleet multi-agent orchestration skill (#505)
Avdhesh Singh Chouhanac53fbcd0ec4
diff --git a/commands/devfleet.md b/commands/devfleet.mdnew file mode 100644index 000000000..7dbef64b9--- /dev/null+++ b/commands/devfleet.md@@ -0,0 +1,92 @@+---+description: Orchestrate parallel Claude Code agents via Claude DevFleet — plan projects from natural language, dispatch agents in isolated worktrees, monitor progress, and read structured reports.+---++# DevFleet — Multi-Agent Orchestration++Orchestrate parallel Claude Code agents via Claude DevFleet. Each agent runs in an isolated git worktree with full tooling.++Requires the DevFleet MCP server: `claude mcp add devfleet --transport http http://localhost:18801/mcp`++## Flow++```+User describes project+ → plan_project(prompt) → mission DAG with dependencies+ → Show plan, get approval+ → dispatch_mission(M1) → Agent spawns in worktree+ → M1 completes → auto-merge → M2 auto-dispatches (depends_on M1)+ → M2 completes → auto-merge+ → get_report(M2) → files_changed, what_done, errors, next_steps+ → Report summary to user+```++## Workflow++1. **Plan the project** from the user's description:++```+mcp__devfleet__plan_project(prompt="<user's description>")+```++This returns a project with chained missions. Show the user:+- Project name and ID+- Each mission: title, type, dependencies+- The dependency DAG (which missions block which)++2. **Wait for user approval** before dispatching. Show the plan clearly.++3. **Dispatch the first mission** (the one with empty `depends_on`):++```+mcp__devfleet__dispatch_mission(mission_id="<first_mission_id>")+```++The remaining missions auto-dispatch as their dependencies complete (because `plan_project` creates them with `auto_dispatch=true`). When manually creating missions with `create_mission`, you must explicitly set `auto_dispatch=true` for this behavior.++4. **Monitor progress** — check what's running:++```+mcp__devfleet__get_dashboard()+```++Or check a specific mission:++```+mcp__devfleet__get_mission_status(mission_id="<id>")+```++Prefer polling with `get_mission_status` over `wait_for_mission` for long-running missions, so the user sees progress updates.++5. **Read the report** for each completed mission:++```+mcp__devfleet__get_report(mission_id="<mission_id>")+```++Call this for every mission that reached a terminal state. Reports contain: files_changed, what_done, what_open, what_tested, what_untested, next_steps, errors_encountered.++## All Available Tools++| Tool | Purpose |+|------|---------|+| `plan_project(prompt)` | AI breaks description into chained missions with `auto_dispatch=true` |+| `create_project(name, path?, description?)` | Create a project manually, returns `project_id` |+| `create_mission(project_id, title, prompt, depends_on?, auto_dispatch?)` | Add a mission. `depends_on` is a list of mission ID strings. |+| `dispatch_mission(mission_id, model?, max_turns?)` | Start an agent |+| `cancel_mission(mission_id)` | Stop a running agent |+| `wait_for_mission(mission_id, timeout_seconds?)` | Block until done (prefer polling for long tasks) |+| `get_mission_status(mission_id)` | Check progress without blocking |+| `get_report(mission_id)` | Read structured report |+| `get_dashboard()` | System overview |+| `list_projects()` | Browse projects |+| `list_missions(project_id, status?)` | List missions |++## Guidelines++- Always confirm the plan before dispatching unless the user said "go ahead"+- Include mission titles and IDs when reporting status+- If a mission fails, read its report to understand errors before retrying+- Agent concurrency is configurable (default: 3). Excess missions queue and auto-dispatch as slots free up. Check `get_dashboard()` for slot availability.+- Dependencies form a DAG — never create circular dependencies+- Each agent auto-merges its worktree on completion. If a merge conflict occurs, the changes remain on the worktree branch for manual resolution.diff --git a/mcp-configs/mcp-servers.json b/mcp-configs/mcp-servers.jsonindex 36ff6bdbc..4efad2ff8 100644--- a/mcp-configs/mcp-servers.json+++ b/mcp-configs/mcp-servers.json@@ -123,6 +123,11 @@},"description": "AI browser agent for web tasks"},+ "devfleet": {+ "type": "http",+ "url": "http://localhost:18801/mcp",+ "description": "Multi-agent orchestration — dispatch parallel Claude Code agents in isolated worktrees. Plan projects, auto-chain missions, read structured reports. Repo: https://github.com/LEC-AI/claude-devfleet"+ },"token-optimizer": {"command": "npx","args": ["-y", "token-optimizer-mcp"],diff --git a/skills/claude-devfleet/SKILL.md b/skills/claude-devfleet/SKILL.mdnew file mode 100644index 000000000..0f4dbd379--- /dev/null+++ b/skills/claude-devfleet/SKILL.md@@ -0,0 +1,103 @@+---+name: claude-devfleet+description: Orchestrate multi-agent coding tasks via Claude DevFleet — plan projects, dispatch parallel agents in isolated worktrees, monitor progress, and read structured reports.+origin: community+---++# Claude DevFleet Multi-Agent Orchestration++## When to Use++Use this skill when you need to dispatch multiple Claude Code agents to work on coding tasks in parallel. Each agent runs in an isolated git worktree with full tooling.++Requires a running Claude DevFleet instance connected via MCP:+```bash+claude mcp add devfleet --transport http http://localhost:18801/mcp+```++## How It Works++```+User → "Build a REST API with auth and tests"+ ↓+plan_project(prompt) → project_id + mission DAG+ ↓+Show plan to user → get approval+ ↓+dispatch_mission(M1) → Agent 1 spawns in worktree+ ↓+M1 completes → auto-merge → auto-dispatch M2 (depends_on M1)+ ↓+M2 completes → auto-merge+ ↓+get_report(M2) → files_changed, what_done, errors, next_steps+ ↓+Report back to user+```++### Tools++| Tool | Purpose |+|------|---------|+| `plan_project(prompt)` | AI breaks a description into a project with chained missions |+| `create_project(name, path?, description?)` | Create a project manually, returns `project_id` |+| `create_mission(project_id, title, prompt, depends_on?, auto_dispatch?)` | Add a mission. `depends_on` is a list of mission ID strings (e.g., `["abc-123"]`). Set `auto_dispatch=true` to auto-start when deps are met. |+| `dispatch_mission(mission_id, model?, max_turns?)` | Start an agent on a mission |+| `cancel_mission(mission_id)` | Stop a running agent |+| `wait_for_mission(mission_id, timeout_seconds?)` | Block until a mission completes (see note below) |+| `get_mission_status(mission_id)` | Check mission progress without blocking |+| `get_report(mission_id)` | Read structured report (files changed, tested, errors, next steps) |+| `get_dashboard()` | System overview: running agents, stats, recent activity |+| `list_projects()` | Browse all projects |+| `list_missions(project_id, status?)` | List missions in a project |++> **Note on `wait_for_mission`:** This blocks the conversation for up to `timeout_seconds` (default 600). For long-running missions, prefer polling with `get_mission_status` every 30–60 seconds instead, so the user sees progress updates.++### Workflow: Plan → Dispatch → Monitor → Report++1. **Plan**: Call `plan_project(prompt="...")` → returns `project_id` + list of missions with `depends_on` chains and `auto_dispatch=true`.+2. **Show plan**: Present mission titles, types, and dependency chain to the user.+3. **Dispatch**: Call `dispatch_mission(mission_id=<first_mission_id>)` on the root mission (empty `depends_on`). Remaining missions auto-dispatch as their dependencies complete (because `plan_project` sets `auto_dispatch=true` on them).+4. **Monitor**: Call `get_mission_status(mission_id=...)` or `get_dashboard()` to check progress.+5. **Report**: Call `get_report(mission_id=...)` when missions complete. Share highlights with the user.++### Concurrency++DevFleet runs up to 3 concurrent agents by default (configurable via `DEVFLEET_MAX_AGENTS`). When all slots are full, missions with `auto_dispatch=true` queue in the mission watcher and dispatch automatically as slots free up. Check `get_dashboard()` for current slot usage.++## Examples++### Full auto: plan and launch++1. `plan_project(prompt="...")` → shows plan with missions and dependencies.+2. Dispatch the first mission (the one with empty `depends_on`).+3. Remaining missions auto-dispatch as dependencies resolve (they have `auto_dispatch=true`).+4. Report back with project ID and mission count so the user knows what was launched.+5. Poll with `get_mission_status` or `get_dashboard()` periodically until all missions reach a terminal state (`completed`, `failed`, or `cancelled`).+6. `get_report(mission_id=...)` for each terminal mission — summarize successes and call out failures with errors and next steps.++### Manual: step-by-step control++1. `create_project(name="My Project")` → returns `project_id`.+2. `create_mission(project_id=project_id, title="...", prompt="...", auto_dispatch=true)` for the first (root) mission → capture `root_mission_id`.+ `create_mission(project_id=project_id, title="...", prompt="...", auto_dispatch=true, depends_on=["<root_mission_id>"])` for each subsequent task.+3. `dispatch_mission(mission_id=...)` on the first mission to start the chain.+4. `get_report(mission_id=...)` when done.++### Sequential with review++1. `create_project(name="...")` → get `project_id`.+2. `create_mission(project_id=project_id, title="Implement feature", prompt="...")` → get `impl_mission_id`.+3. `dispatch_mission(mission_id=impl_mission_id)`, then poll with `get_mission_status` until complete.+4. `get_report(mission_id=impl_mission_id)` to review results.+5. `create_mission(project_id=project_id, title="Review", prompt="...", depends_on=[impl_mission_id], auto_dispatch=true)` — auto-starts since the dependency is already met.++## Guidelines++- Always confirm the plan with the user before dispatching, unless they said to go ahead.+- Include mission titles and IDs when reporting status.+- If a mission fails, read its report before retrying.+- Check `get_dashboard()` for agent slot availability before bulk dispatching.+- Mission dependencies form a DAG — do not create circular dependencies.+- Each agent runs in an isolated git worktree and auto-merges on completion. If a merge conflict occurs, the changes remain on the agent's worktree branch for manual resolution.+- When manually creating missions, always set `auto_dispatch=true` if you want them to trigger automatically when dependencies complete. Without this flag, missions stay in `draft` status.