review-code-quality
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions mandate a minimum findings quota of three issues per analyzer, requiring a 200-character justification if none are found. While intended to enforce rigor, this behavioral constraint can pressure the agent to highlight minor or non-existent issues.
- [COMMAND_EXECUTION]: The skill performs shell commands including 'git diff' to detect modified files and uses the 'wc' utility via a restricted Bash tool in the readability agent. These operations are limited to the skill's primary function of code analysis.
- [INDIRECT_PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it ingests and processes the content of source code files that may contain adversarial instructions or malicious comments.
- Ingestion points: Modified files are identified via 'git diff' and then processed by multiple sub-agents using the Read tool.
- Boundary markers: No specific delimiters or safety instructions are used to distinguish file content from the agent's core instructions.
- Capability inventory: The skill set includes file system access (Read, Grep, Glob), restricted shell execution (wc), and git operations.
- Sanitization: There is no evidence of content sanitization or escaping before the code is interpolated into the agent's context.
Audit Metadata