thermos
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from external sources including user requests, pull requests, and file diffs, which creates a surface for instructions embedded in data to influence subagent behavior.\n
- Ingestion points: The workflow in
SKILL.mdidentifies ingestion of data from user requests, PRs, current branches, and changed files.\n - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' prompts to isolate the code under review from the subagents' operational instructions.\n
- Capability inventory: The skill has the capability to launch multiple subagents (
thermo-nuclear-review-subagentandthermo-nuclear-code-quality-review-subagent) and synthesize their output.\n - Sanitization: No evidence of sanitization, escaping, or validation of the ingested diffs or context excerpts is provided.
Audit Metadata