code-investigator
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions define a professional debugging workflow that relies on local environment tools and evidence-based hypothesis testing. It explicitly directs the agent to identify security concerns such as authentication bypass and injection during its investigations.
- [PROMPT_INJECTION]: The skill processes untrusted input in the form of logs, stack traces, and error messages, which creates a surface for indirect prompt injection. This is an inherent risk for diagnostic tools and is mitigated by the skill's requirement for evidence-based confirmation.
- Ingestion points: Workflow Step 1 in
SKILL.md(reading logs, stack traces, and bug descriptions). - Boundary markers: No specific delimiters are defined for isolating untrusted log content from instructions.
- Capability inventory: Access to repository files (git diff/status), search tools, and local execution of test/lint commands.
- Sanitization: No explicit sanitization of log content is mandated.
Audit Metadata