Defines governance metrics catalogue, snapshot formats, staleness thresholds, and audit report specs for tracking project governance health in audits and dashboards.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ai-literacy-superpowers:governance-observabilityThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Defines the metrics, snapshot formats, and dashboard specifications
Defines the metrics, snapshot formats, and dashboard specifications for governance health. This skill is referenced by:
governance-auditor agent when producing audit reports/governance-health command when displaying results/governance-audit command when updating snapshots| Metric | Type | Description |
|---|---|---|
| Total constraints | integer | Total governance constraints in HARNESS.md |
| Falsifiability ratio | percentage | Proportion of governance constraints scored as "falsifiable" |
| Falsifiable | integer | Number of constraints rated "Falsifiable" |
| Vague | integer | Number of constraints rated "Vague" |
| Semantic drift stage | integer (1-5) | Numeric semantic drift severity from the five-stage model |
| Drift velocity | enum (stable/increasing/decreasing) | Direction of drift trend between audits |
| Governance debt items | integer | Number of governance debt items identified |
| Aggregate debt score | integer | Sum of (severity x blast_radius) across all debt items |
| Frame alignment score | percentage | Proportion of governance constraints with confirmed three-frame alignment |
Governance metrics are expressed as a markdown section in audit reports. The governance-auditor includes a Governance Summary section that agents (including the snapshot generator) can parse directly:
## Governance Summary
- Total constraints: 4
- Falsifiable: 3 (with verification criteria)
- Vague: 0 (lacking operational meaning)
- Falsifiability ratio: 75%
- Semantic drift stage: 1/5
- Drift velocity: stable
- Governance debt items: 2
- Aggregate debt score: 8 (sum of severity x blast radius)
- Frame alignment score: 50%
| Metric | Fresh | Stale | Critical |
|---|---|---|---|
last_audit | < 90 days | 90-180 days | > 180 days |
drift_velocity | stable or decreasing | increasing for 1 quarter | increasing for 2+ quarters |
falsifiability_ratio | > 0.75 | 0.50-0.75 | < 0.50 |
debt_inventory_size | 0-2 items | 3-5 items | > 5 items |
Governance audit reports are written to
observability/governance/audit-YYYY-MM-DD.md by the
governance-auditor agent. Each report contains:
See references/dashboard-spec.md for the full HTML dashboard
specification — sections, data sources, visualisation types, and
colour schemes.
When governance data is aggregated across repos (via
/portfolio-assess), the portfolio dashboard gains:
npx claudepluginhub habitat-thinking/ai-literacy-superpowers --plugin ai-literacy-superpowersGuides governance audits: detects semantic drift in constraints, inventories governance debt, scores falsifiability, checks three-frame alignment for governance-auditor agent.
Encodes human-readable governance policies into machine-executable JSON constraints for AI agents and CI pipelines to validate automatically. Outputs rule files in .ai/governance/.
Pulls snapshots from configured metric sources like GitHub, computes deltas against prior snapshots, flags unexplained signals, and drafts evidence entries for canvas files. Use before assessments, after launches, or weekly.