smart-code-review
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local command-line utilities to scan the project structure and analyze code changes. Specifically, it uses
gitcommands (git status,git diff) to determine review scope and search tools likerg(ripgrep),grep, andfindto locate specific files or code entities during targeted reviews. - [SAFE]: No data exfiltration or unauthorized network operations were identified. The skill operates entirely on the local file system to provide insights and does not transmit sensitive code or credentials to external servers.
- [SAFE]: The skill includes a built-in safety mechanism that prevents the agent from making autonomous modifications to the codebase. It requires a mandatory 'Next Steps' confirmation from the user before implementing any proposed fixes.
- [SAFE]: The skill processes project source code which is an untrusted ingestion surface (Indirect Prompt Injection); however, the risk is mitigated by the review-only nature of the tool and the requirement for human oversight before any actions are taken.
Audit Metadata