coderabbit
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill demonstrates a strong security posture by providing explicit warnings about verifying installation binaries and checksums. It also advises on using minimum permission scopes for authentication.
- [EXTERNAL_DOWNLOADS]: Directs users to the official CodeRabbit website (coderabbit.ai) for CLI installation. It specifically cautions against unsafe installation patterns such as piping remote scripts directly into a shell.
- [PROMPT_INJECTION]: Addressed through the identification of indirect prompt injection surfaces from external code and tool output. 1. Ingestion points: Local repository content and CodeRabbit review output (SKILL.md). 2. Boundary markers: The skill includes a 'Security note' specifically warning the agent to treat these sources as untrusted. 3. Capability inventory: Use of shell commands via the
coderabbitandgitCLIs (SKILL.md). 4. Sanitization: Relies on explicit instructional constraints to prevent the agent from executing code found within review results. - [COMMAND_EXECUTION]: Uses standard CLI tools (
coderabbit,git) for its core review and authentication functions. The commands are scoped to the intended purpose of the skill.
Audit Metadata