thermo-nuclear-code-quality-review
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data (source code changes) that could contain malicious instructions designed to override the agent's review logic or influence its output.
- Ingestion points: The skill is designed to analyze the "current branch's changes" and project diffs (SKILL.md).
- Boundary markers: Absent. The instructions do not provide delimiters or "ignore embedded instructions" warnings to isolate the code being audited from the agent's reasoning process.
- Capability inventory: The skill writes multiple markdown reports to the local file system (
/tmp/<project>-thermo/) and describes a workflow involving the execution of verification commands and coordination with a 'swarm' skill. - Sanitization: Absent. There is no mention of filtering or escaping instructions found within the audited code.
- [COMMAND_EXECUTION]: The skill instructions require the agent to generate reports that include "measurements" and "the commands that produced them," implying the execution of shell-based diagnostic or auditing tools.
- Evidence: The "Report Format" section specifies that detail files should carry measurements and the commands used to generate them (SKILL.md).
Audit Metadata