w-docs
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository (code and existing documentation) and has the capability to write or patch sensitive files, including agent rules. This creates an attack surface where malicious instructions hidden in the code could be propagated into the agent's internal constraints.
- Ingestion points: The skill reads repository files, architecture docs (referenced in AGENTS.md), and READMEs to identify changes.
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions within the ingested code or documentation.
- Capability inventory: The skill is authorized to perform file-write and patch operations on .md, .mdx, and rule files.
- Sanitization: There is no evidence of sanitization or validation of repository content before it is interpolated into documentation or rule patches.
Audit Metadata