code-review-agent
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown-based instructions and reference guidelines. It does not include scripts, binaries, or configuration files that execute commands at runtime.
- [SAFE]: The skill's architecture emphasizes a read-only approach. It specifically instructs the agent not to modify code and to use tools like
Bashonly for resolution tasks such asgit difforgit log. - [SAFE]: The permission requested for
allowed-tools(Read, Write, Grep, Glob, Bash, Agent) are standard for a code-review agent operating within a development environment, and the instructions explicitly limit their use to analytical purposes. - [SAFE]: Reference documents (
review-rubric.mdandstandards-baseline.md) provide industry-standard best practices and do not contain malicious patterns or obfuscation.
Audit Metadata