thermo-nuclear-review
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is instructed to audit branch changes and review pull request discussions, which are ingestion points for untrusted data that could contain malicious instructions designed to override the agent's behavior.
- Ingestion points: The prompt in SKILL.md directs the agent to analyze the diff of a checked-out branch and to fetch discussion content via the
ghorglabCLI tools. - Boundary markers: There are no explicit delimiters or specific 'ignore' instructions provided to the agent to distinguish between its internal logic and the untrusted content it is auditing.
- Capability inventory: The agent uses standard file system access to read code diffs and utilizes the
ghandglabCLI tools to retrieve external discussion data from GitHub or GitLab. - Sanitization: The instructions do not include any requirements for sanitizing, escaping, or validating the external content before it is processed as part of the audit.
Audit Metadata