code-review
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions and scripts recommend the installation of the
@coderabbitai/clitool from the npm registry to enable its core functionality. - [COMMAND_EXECUTION]: The skill utilizes shell scripts to execute
gitcommands and thecoderabbitCLI to perform code analysis and manage review metadata. - [INDIRECT_PROMPT_INJECTION]: The skill reads and processes feedback generated by the external CodeRabbit AI service. This creates a surface where malicious content within the code being reviewed could influence the AI's output, potentially leading the agent to follow embedded instructions.
- Ingestion points: Review results saved as Markdown files in the
.ada/data/reviews/directory. - Boundary markers: None identified; the agent is instructed to read and address the issues found in the raw review content.
- Capability inventory: File system read/write access, execution of git commands, and network communication through the CodeRabbit CLI.
- Sanitization: No evidence of output sanitization or filtering before the review content is presented to the agent.
Audit Metadata