maintain-repository-docs
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted files from a repository to generate documentation, creating a vulnerability surface where malicious code in the repository could influence agent behavior.
- Ingestion points: The skill reads manifests, source code, CI configurations, and scripts as evidence (SKILL.md).
- Capability inventory: The agent is authorized to execute shell commands and run examples found in the repository (references/structure-and-validation.md).
- Boundary markers: There are no explicit instructions for the agent to treat repository content as untrusted data or use protective delimiters.
- Sanitization: No sanitization or validation steps are defined for the content extracted from the repository before it is processed or executed.
- [DYNAMIC_EXECUTION]: The skill provides instructions to execute code and binary files discovered within the target repository.
- Evidence: references/structure-and-validation.md directs the agent to "Run changed examples" and "confirm option names using its actual parser/help". While intended for verification, this involves executing arbitrary code from an external source.
Audit Metadata