thermo-nuclear-review
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from git diffs, file contents, and PR/MR discussions, which presents a surface for indirect prompt injection attacks.
- Ingestion points: As described in
SKILL.mdandreferences/thermo-nuclear-review-subagent.md, the agent ingests external data including git diff outputs and PR comments retrieved throughghorglabCLI tools. - Boundary markers: Input data is organized into labeled markdown sections (e.g.,
### Git / diff output), providing structural delimiters for the agent. - Capability inventory: The skill performs read operations on repository files and PR discussions. It does not possess capabilities for arbitrary file writing or uncontrolled network communication.
- Sanitization: The instructions do not specify explicit sanitization or filtering for the external content, relying on the agent's auditing role to identify risks rather than execute content.
Audit Metadata