interrogate
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as pull request descriptions, commits, and branch diffs to determine its intent and scope.
- Ingestion points: Instructions in SKILL.md direct the agent to read the branch diff, commits, and PR description.
- Boundary markers: There are no explicit delimiters or instructions provided to differentiate between the agent's instructions and the content of the data it is analyzing.
- Capability inventory: The skill requires the agent to prove each finding with a run against the real code using repro scripts or command output, which requires shell execution capabilities.
- Sanitization: There is no mention of sanitizing or validating the contents of the diffs or descriptions before processing or executing commands based on them.
- [DYNAMIC_EXECUTION]: The skill instructs the agent to generate and execute reproduction scripts or commands to verify findings against the codebase. While this is the core intended purpose for adversarial testing, it involves the dynamic creation of executable content at runtime.
Audit Metadata