The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the development environment and external sources.
Ingestion points: The skill reads git diff output, git log entries, repository pull request templates, and existing PR comments via gh api calls in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or "ignore" instructions for the ingested content to distinguish it from the agent's core instructions.
Capability inventory: The skill possesses the ability to write back to the environment via git push, gh pr create, gh pr edit, and gh api (POST) to reply to comments.
Sanitization: While the skill enforces a strict "voice contract" (concise, human-readable bullets) which limits the complexity of the output, there is no technical sanitization or filtering of the raw input data.
[COMMAND_EXECUTION]: The skill utilizes the git and gh (GitHub CLI) command-line tools to interact with repositories and pull requests.
It performs repository operations such as git diff, git log, and git push.
It interacts with the GitHub API and CLI for pull request metadata and comment management.

concise