The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection from untrusted pull request data. Ingestion points: Pull request titles, descriptions, and git diffs are processed in steps 3 and 4 of SKILL.md. Boundary markers: There are no explicit delimiters or instructions provided to the agents to disregard embedded commands in the PR content. Capability inventory: The skill has the ability to post comments and create inline suggestions on GitHub via the gh CLI and dedicated MCP tools. Sanitization: There is no logic present to sanitize or escape the content retrieved from the pull request before it is processed by the review agents.
[COMMAND_EXECUTION]: The skill uses the gh command-line tool for essential operations including viewing pull requests, retrieving git history, and posting review comments.
[EXTERNAL_DOWNLOADS]: The documentation references the installation of the official GitHub CLI, which is a well-known service.

code-review