# Microgram: flywheel-mirror

`station__algovigilance__microgram-flywheel-mirror` · native (always callable) · domain `algovigilance` · pv-relevance `pv-core`

Self-observation gate for the brain flywheel. Takes brain health metrics and determines if the flywheel can see itself spinning. Checks implicit store health, verdict coverage (sigma), tracker integrity, and guardian responsiveness. GREEN = full self-observation, YELLOW = partial (degraded tools), RED = blind (cannot measure own state).

## Agent metadata

- `idempotent`: true
- `read_only`: true
- `expected_latency_ms`: unknown (not yet contract-tested)
- `cost_tokens_estimate`: unknown

## Input schema

- `implicit_healthy` *boolean* — Whether implicit_stats tool returns successfully
- `verdict_pct` *number* — Percentage of sessions with model-authored verdicts (0-100)
- `tracker_ok` *boolean* — Whether brain tracker index parses without error
- `guardian_ok` *boolean* — Whether guardian_homeostasis_tick completes

## Example call

```json
POST /api/mcp
Content-Type: application/json

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "station__algovigilance__microgram-flywheel-mirror",
    "arguments": {}
  }
}
```

## Related

- [/tools](/tools) — all 7718 tools
- [/tools/algovigilance__microgram-flywheel-mirror](/tools/algovigilance__microgram-flywheel-mirror) — HTML page
- [/tools/algovigilance__microgram-flywheel-mirror/json](/tools/algovigilance__microgram-flywheel-mirror/json) — JSON form (agent-friendly)
- [/api/mcp](/api/mcp) — endpoint
- [/AGENTS.md](/AGENTS.md) — agent guide