phdev/Depthfinder context-honesty gate
Fail CI when your AI context files (CLAUDE.md / AGENTS.md / .cursorrules) drift out of sync with the code.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| path | Directory to scan (default: repo root). | no | . |
| max-false | Allow up to N false Context claims before failing (default 0). | no | 0 |
| warn-below | Also fail if any Health dimension (Honesty/Weight/Coverage/Health) scores below N (0-100). Empty = off. | no | "" |
| weight-budget | Token budget for the Weight dimension (heuristic). Empty = default (10000). | no | "" |
| docs | Also scan the wider repo docs for an advisory Doc Honesty score (never gated). true/false. | no | false |
| version | depthfinder npm version to pin. Pinning keeps the gate stable across releases. | no | 1.3.0 |
Outputs
no outputs