code-review
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (
git diff,git show) using direct user input for commit hashes, branch names, and ranges in Step 0. If these inputs are not properly sanitized by the agent, they could lead to command argument injection or execution of unintended shell operations. - [PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection (Category 8) as it processes untrusted data from git repositories. Malicious instructions could be embedded in code comments or PR descriptions to override the agent's review logic.
- Ingestion points: The agent reads untrusted code content from the filesystem using
gitcommands specified inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when interpolating the diff content into the analysis prompt.
- Capability inventory: The skill has access to
bashfor executing git commands and broad filesystem read access for stack detection and code analysis. - Sanitization: The skill does not perform any sanitization or validation of the code content before it is processed by the agent.
Audit Metadata