general-code-reviewing
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data including pull requests, branch diffs, and source files. Maliciously crafted code or comments within these targets could attempt to influence the agent's review logic. Ingestion points: Processes PRs, branch diffs, staged diffs, commit ranges, and named files (SKILL.md). Boundary markers: The instructions do not specify any delimiters or 'ignore embedded instructions' warnings for the data being reviewed. Capability inventory: The skill coordinates the execution of
adversarial-code-reviewingandsimplicity-reviewskills to analyze code and return synthesized reports. Sanitization: There is no evidence of sanitization, filtering, or escaping for the ingested code content. - [NO_CODE]: The skill consists entirely of markdown instructions and does not include any scripts, executables, or configuration files.
Audit Metadata