doc
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads and processes content from the repository (e.g., code, documentation, git logs) to generate new documentation. This introduces an attack surface where malicious instructions embedded in the repository files could influence the agent's output.\n
- Ingestion points: Repository files (README, source code, config), git history, and existing documentation.\n
- Boundary markers: The skill contains instructions to "reuse accepted intent," distinguish between observed facts and unverified claims, and "ignore unverified claims."\n
- Capability inventory: The skill can write files to the repository (documentation, handoffs, metadata) and execute a local audit shell script.\n
- Sanitization: Explicit instructions are provided to "redact secrets and unrelated personal data" when generating documentation.\n- [COMMAND_EXECUTION]: The skill executes a bundled shell script (
scripts/audit-oss-docs.sh) to audit documentation presence. This script performs local filesystem operations such asfind,grep, andbasenameto detect project types and file presence. It does not perform network operations or access sensitive system directories.
Audit Metadata