refactor-pipeline
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository and external review threads, which could potentially contain malicious instructions intended to influence the agent's behavior.\n
- Ingestion points: The skill reads source code for refactoring and monitors GitHub PR review threads for feedback (Shepherd lane).\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the ingested content.\n
- Capability inventory: The agent performs git operations (pushing branches), GitHub CLI operations (creating PRs via
gh stack), and executes arbitrary test commands defined within the project.\n - Sanitization: No sanitization or validation of the ingested code or review comments is performed before processing.\n- [COMMAND_EXECUTION]: The instructions direct the agent to execute project-defined tests to validate refactors (e.g., 'run the tests for the affected crates'). This allows for the execution of commands specified in the project's build and test configuration.
Audit Metadata