receiving-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes feedback from external reviewers, which constitutes untrusted data that could contain malicious instructions or suggestions to weaken codebase security.
- Ingestion points: External review feedback and human partner instructions processed in SKILL.md.
- Boundary markers: No specific technical delimiters are used to wrap the input data; the skill relies on the agent's logic to verify content.
- Capability inventory: The skill allows for codebase searching (grep), GitHub API interactions (gh api), and implementation of code changes.
- Sanitization: The workflow requires manual technical evaluation and verification against the codebase as a primary defense.
- [COMMAND_EXECUTION]: The skill guides the agent to use command-line utilities to inspect the local environment and interact with GitHub.
- Evidence: Examples provided for grep usage to verify code presence and gh api for replying to PR threads (SKILL.md).
Audit Metadata