| name | flywheel-healthcheck |
| description | Full health check of the codebase and flywheel dependencies. |
Run a health check. $ARGUMENTS
Produce a health score across dependencies, codebase state, and beads.
Dependency checks (run in parallel via Bash):
br --version — bead tracker
bv --version — bead visualizer
curl -s --max-time 3 http://127.0.0.1:8765/health/liveness — agent-mail
git status --short — repo cleanliness
ls .pi-flywheel/checkpoint.json 2>/dev/null — checkpoint state
ls mcp-server/dist/server.js 2>/dev/null — MCP server built
Codebase health (use Agent(Explore)):
- TODO/FIXME count:
grep -r "TODO\|FIXME\|HACK" --include="*.ts" | wc -l
- Test ratio: count test files vs source files
- Any obvious linting or compilation errors
Bead health:
- Run
br list --json — count open/closed/in-progress/deferred
- Run
bv --json — check for cycles or orphaned beads
Display health report:
DEPENDENCIES
✅ br v1.x.x
✅ bv v1.x.x
✅ agent-mail — healthy
⚠️ MCP server not built (run: cd mcp-server && npm run build)
CODEBASE
TODOs: N
Test coverage: N% (N test files / N source files)
Git: clean / N uncommitted changes
BEADS
Open: N | In progress: N | Closed: N | Deferred: N
Graph: ✅ no cycles | ⚠️ N orphans
HEALTH SCORE: N/10
See also (triage chain)
Healthcheck is the third of three diagnostic commands. Run them in order:
/agent-flywheel:flywheel-doctor — read-only snapshot. Run first when toolchain drift is suspected.
/agent-flywheel:flywheel-setup — apply-fixes stage; installs tools, registers MCP, configures hooks.
/agent-flywheel:flywheel-healthcheck (this skill) — deep periodic audit of codebase, bead graph, and dependencies. Run it on a cadence, not for fresh-clone setup problems (use doctor → setup for that).