The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill documentation and installation guides (e.g., in SKILL.md and references/cli-usage.md) provide commands to install the CodeRabbit CLI via curl -fsSL https://cli.coderabbit.ai/install.sh | sh. This involves piping a remote script to the shell. The resource originates from the official domain of CodeRabbit, which is a recognized provider of AI developer tools.
[COMMAND_EXECUTION]: The helper script scripts/run_coderabbit.py uses the subprocess module to execute git and the coderabbit CLI. This functionality is essential for the skill's primary purpose of automating code reviews and verifying project states.
[PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface as it instructs agents to process and implement fixes based on output generated by the external CodeRabbit CLI.
Ingestion points: Review findings are captured in coderabbit-report.txt and subsequently read by the AI agent.
Boundary markers: The skill lacks specific delimiters or instructional guardrails to prevent the agent from obeying potentially malicious instructions embedded in the external report.
Capability inventory: The agent is granted capabilities to modify the codebase and execute commands based on the triaged findings.
Sanitization: There is no evidence of sanitization or validation of the external output before it is processed by the agent.

coderabbit