investigate
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes common development utilities such as
grep,git,npm,pnpm, andpytestto analyze codebase history and run test suites. These operations are essential for identifying and verifying the root cause of reported issues. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied issue descriptions which could theoretically contain malicious instructions.
- Ingestion points: Untrusted data enters the agent context through the
<issue_description>argument inSKILL.md. - Boundary markers: No explicit delimiters are used to wrap the injected issue description, though the instructions guide the agent to treat it as data to be analyzed.
- Capability inventory: The agent has the ability to search files, view git history, and execute local test commands.
- Sanitization: Input from the issue description is not sanitized before being used to form search queries or debugging hypotheses.
Audit Metadata