kming-l/LyingDocs
Documentation-code misalignment detection. A trust layer for your repository.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| doc-path | Path to documentation root directory (relative to repo root) | yes | docs |
| code-path | Path to code repository root (relative to repo root) | no | . |
| config | Path to lyingdocs.toml config file (optional) | no | "" |
| backend | Argus backend: local, claude_code, or codex | no | local |
| hermes-provider | LLM provider for Hermes agent: openai or anthropic. Auto-set to 'anthropic' when backend is claude_code. | no | "" |
| argus-provider | LLM provider for Argus local backend: openai or anthropic | no | "" |
| hermes-model | LLM model for Hermes agent | no | "" |
| hermes-base-url | API base URL for Hermes agent | no | "" |
| argus-model | LLM model for Argus agent | no | "" |
| argus-base-url | API base URL for Argus agent | no | "" |
| openai-api-key | OpenAI API key (for local/codex backends) | no | "" |
| anthropic-api-key | Anthropic API key (for claude_code backend, billed per API usage) | no | "" |
| claude-oauth-token | Claude Code OAuth token (for claude_code backend, uses Pro/Max subscription quota). Generate with: claude setup-token | no | "" |
| max-dispatches | Maximum number of Argus dispatches | no | "" |
| max-iterations | Maximum Hermes loop iterations | no | "" |
| gen-issue | Generate GitHub issue drafts from findings (true/false) | no | false |
| comment-on-pr | Post findings as a PR comment (true/false) | no | true |
| python-version | Python version to use | no | 3.12 |
Outputs
| name | description |
|---|---|
| report-path | Path to the generated Misalignment Report |
| findings-count | Number of misalignment findings |
| has-findings | Whether any misalignments were found (true/false) |