The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard development tools such as git (status, diff), gh (GitHub CLI), and rg (ripgrep) to evaluate code changes and retrieve repository information.
[COMMAND_EXECUTION]: The workflow includes instructions to execute existing project-defined test commands and package scripts to verify implementation changes during simplification or fix modes.
[PROMPT_INJECTION]: The skill processes untrusted input in the form of pull request diffs and external project rules, which represents a potential surface for indirect prompt injection. This is a common characteristic of code review tools and is mitigated by the skill's explicit instruction to the agent to identify security and injection risks within the reviewed code. * Ingestion points: Pull request content retrieved via gh pr diff and git diff, along with local project files like AGENTS.md and CLAUDE.md. * Boundary markers: The instructions do not define explicit delimiters to separate instructions from untrusted data in diff hunks. * Capability inventory: Shell command access for standard development utilities and project-specific verification scripts. * Sanitization: No validation or filtering is applied to the diff content before processing.

pr-review-toolkit