The Agent Skills Directory

[SAFE]: The skill consists of markdown instructions and configuration for pull request review logic. It does not include any scripts, command execution, or network operations.
[SAFE]: The skill contains a reference to a configuration file on the vendor's GitHub repository: github.com/thrillmade/agent-skills/blob/main/docs/skill-versions.json. This is a standard practice for version tracking within the vendor's ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data in the form of code diffs from pull requests. While this presents an attack surface where a malicious diff could contain instructions for the AI agent, the skill's capabilities are limited to providing review comments. It does not possess high-risk tools like file writers or shell execution that could be exploited.

test-discipline