The Agent Skills Directory

Command Execution: The skill is designed to run validation scripts, tests, and linters defined within the local repository to verify pull request readiness. Although it aims to use non-mutating and focused checks, it relies on the integrity of the repository's configuration and scripts.
Indirect Prompt Injection: The skill ingests data from the repository, such as commit history, diffs, and pull request templates, to generate descriptions. There is a potential for untrusted content within these sources to influence the agent's behavior during the pull request preparation process.
File Modification and Committing: The skill has the capability to apply repairs and create commits when authorized by the user to fix validation failures. This involves the agent modifying source code and interacting with the local git history based on pre-defined repository conventions.

pull-request