flywheel-setup

name	flywheel-setup
description	Set up flywheel prerequisites for this project.

Set up the agent-flywheel for this project. $ARGUMENTS

Check and configure all prerequisites:

1. MCP server (build): Locate and verify the agent-flywheel MCP server is built.

   • The plugin's skills/ directory is typically a symlink to the repo root's skills/ folder. Resolve it to find mcp-server/ as a sibling:

     SKILLS_REAL=$(readlink ~/.claude/plugins/marketplaces/local-desktop-app-uploads/agent-flywheel/skills 2>/dev/null)
     MCSRV="${SKILLS_REAL%/skills}/mcp-server"
     test -f "$MCSRV/dist/server.js" && echo "OK: $MCSRV" || echo "MISSING: $MCSRV"
     

   • If MISSING: STOP. Instruct:

     cd <resolved-mcp-server-path> && npm install && npm run build
     

   • Do not proceed to other checks until this passes.

2. MCP server (registered): Verify the MCP server is actually loaded in this Claude Code session.

   • Run ToolSearch("flywheel_profile") to check if flywheel tools are available.
   • If no results: the server is built but not registered. Instruct: "Add the agent-flywheel server entry to your Claude Code MCP config (.mcp.json in the plugin dir or claude_desktop_config.json) and restart Claude Code."

3. br CLI: Run br --version via Bash.

   • If not found: "br is not installed. Install from https://github.com/burningportra/br"
   • If found: check .beads/ directory. If missing, offer to run br init.

4. bv CLI: Run bv --version via Bash. Report status.

5. agent-mail: Test curl -s --max-time 3 http://127.0.0.1:8765/health/liveness via Bash.

   • If reachable: call health_check via agent-mail MCP tool.
   • If not reachable: "agent-mail is not running. Start it with: uv run python -m mcp_agent_mail.cli serve-http"

6. Pre-commit guard: Call install_precommit_guard via agent-mail MCP tool with project_key and code_repo_path set to the current working directory.

7. DCG (Destructive Command Guard): Check if a PreToolUse hook exists in the project's .claude/settings.json that blocks destructive commands. If not, create one:

   {
     "hooks": {
       "PreToolUse": [
         {
           "matcher": "Bash",
           "hooks": [
             {
               "type": "command",
               "command": "bash -c 'echo \"$CLAUDE_TOOL_INPUT\" | jq -r .command | grep -qiE \"(rm\\s+-rf|git\\s+reset\\s+--hard|git\\s+clean\\s+-f|git\\s+checkout\\s+\\.\\s|git\\s+push\\s+--force|drop\\s+table|truncate\\s+table)\" && echo \"BLOCKED: Destructive command detected. Ask the user for explicit permission.\" && exit 1 || exit 0'"
             }
           ]
         }
       ]
     }
   }
   

   If .claude/settings.json already exists with other hooks, merge the PreToolUse entry rather than overwriting. This provides mechanical enforcement beyond the social rules in AGENTS.md.

8. Register agent: Call register_agent via agent-mail MCP tool with project_key and agent_name: "Orchestrator".

9. Display a health checklist:

   ✅ MCP server built (dist/server.js exists)
   ✅ MCP server registered (flywheel_profile tool available)
   ✅ br v1.x.x — beads initialized
   ✅ bv v1.x.x
   ✅ agent-mail — healthy
   ✅ pre-commit guard installed
   ✅ DCG hook active
   ✅ agent registered as "Orchestrator"
   

10. Gate recommendation:

• If ALL checks pass: "All prerequisites met. You can now run /flywheel."
• If ANY check failed: "Do not run /flywheel until all checks pass. Fix the items marked ❌ above and re-run /flywheel-setup."