mergepatrol
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of
verifycommands defined within the repository's.mergepatrol/mergepatrol.ymlfile during the automated merge process.- [INDIRECT_PROMPT_INJECTION]: The skill relies on external data sources that represent potential injection surfaces: It monitors pull request labels and comments to manage queue status, and it uses an agent to resolve code conflicts and automatically generate commits to fix CI failures based on build logs, creating a scenario where malicious content in logs could influence agent behavior.- [DYNAMIC_EXECUTION]: The 'fix_ci' feature allows an agent to dynamically generate and commit new code changes to the repository to resolve build errors, which constitutes automated runtime code generation.
Audit Metadata