code-review-skill
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform analysis using command-line tools such as git and grep to map the project and trace symbol usage.
- [DATA_EXPOSURE]: The skill incorporates security best practices by explicitly excluding sensitive directories and file types, such as .env files, private keys, and credentials, from the audit process.
- [INDIRECT_PROMPT_INJECTION]: As the skill is designed to read and process external codebase data, it possesses an inherent surface for indirect prompt injection. 1. Ingestion points: Source files identified during codebase exploration. 2. Boundary markers: No specific delimiters are defined to separate untrusted code from agent instructions. 3. Capability inventory: Access to file reading and shell commands (git, grep). 4. Sanitization: No explicit content sanitization is described.
Audit Metadata