thermo-nuclear-code-quality-review
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve analyzing external, potentially untrusted code from branch changes. This introduces a surface for indirect prompt injection if the analyzed code contains malicious instructions targeting the agent.
- Ingestion points: SKILL.md (Logic for auditing branch changes).
- Boundary markers: The prompt lacks explicit delimiters to isolate the code under review from the agent's instructions.
- Capability inventory: The skill directs output to the file system via specified report paths in the
/tmp/directory. - Sanitization: The skill does not define methods to sanitize or filter content from the ingested code.
Audit Metadata