What problem does it solve?
This Skill helps prevent inaccurate, stale, or hallucinated technical documentation from reaching users. It treats documentation as a set of verifiable claims and checks those claims against the actual codebase, configuration, routes, flags, and examples.
Core Features & Use Cases
- Documentation Fact-Checking: Verifies symbols, function names, CLI flags, endpoints, config keys, file paths, and behavior claims against source code.
- Code Sample Validation: Reviews examples for real imports, correct signatures, realistic placeholders, and runnable setup on a clean machine.
- Cross-Surface Drift Detection: Checks whether READMEs, references, docstrings, changelogs, and config samples stay consistent after code changes.
- Review Reporting: Produces structured findings with rule-based severity and file:line evidence for false or unverifiable claims.
- Use Case: After an AI updates a README and API docstrings for a feature rename, use this Skill to confirm every referenced symbol exists, examples still work, and old names were removed from all documentation surfaces.
Quick Start
Ask the agent to use the docs-guard skill to review the changed README and docstrings against the current source before publishing.