receiving-thermo-review
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources which could contain malicious instructions designed to influence the agent's code-editing behavior.
- Ingestion points: According to Workflow Step 1 in
SKILL.md, the skill ingests "thermo-review" reports, PR comments, and unstructured feedback. - Boundary markers: The instructions do not specify the use of delimiters or specific commands to ignore potential instructions embedded within the ingested feedback.
- Capability inventory: The skill is designed to perform file modifications and apply code fixes as outlined in the "Scope and Authorization" and "Item Handling" sections of
SKILL.md. - Sanitization: The skill incorporates protective measures such as the "Regression Guard" and a "Push Back" framework, which require the agent to verify all findings against current code behavior and architectural intent before applying changes.
Audit Metadata