design-doc-reviewer
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows a well-defined, structured workflow to read documentation and generate reviews. It does not use any network tools, external scripts, or elevated privileges.
- [DATA_EXPOSURE]: The skill reads project documentation from 'docs/design/' and 'docs/vision/vision.md'. These file access operations are limited to the local filesystem and are consistent with the skill's stated purpose.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: Reads arbitrary content from user-specified design documents and files in the 'docs/design/' directory.
- Boundary markers: Absent. The skill does not wrap the ingested document content in specific delimiters or instruct the agent to ignore instructions within the files.
- Capability inventory: File read access (documentation) and file write access (saving reviews to 'docs/design/').
- Sanitization: Absent. The skill does not explicitly sanitize the content of the documents before processing. The risk is mitigated by the highly structured output format and the specific scorecard rubric which constrains the agent's response logic.
Audit Metadata