coderabbit
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill downloads and executes an installation script from the official CodeRabbit domain (
https://cli.coderabbit.ai/install.sh). This is the intended installation method for the CLI tool integrated by this skill. - [COMMAND_EXECUTION]: The skill utilizes shell commands for tool authentication (
coderabbit auth login) and triggering reviews (coderabbit --prompt-only). These operations are essential for the integration's functionality. - [INDIRECT_PROMPT_INJECTION]: The skill processes project source code and linter outputs to generate reviews, representing an indirect prompt injection surface.
- Ingestion points: CLI output from
coderabbitand project source files mentioned inSKILL.md. - Boundary markers: The instructions specify using the
--prompt-onlyflag to provide AI-optimized, structured feedback to the agent. - Capability inventory: The skill requires shell execution and local file system access.
- Sanitization: Relies on the CodeRabbit CLI tool to filter and format analysis results before they are consumed by the AI agent.
Audit Metadata