The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the git and gh (GitHub CLI) command-line tools to perform repository operations including logging changes (git log), checking differences (git diff), tagging releases (git tag), and pushing updates to remote servers (git push).
[INDIRECT_PROMPT_INJECTION]: The workflow relies on data extracted from the project's Git history, such as commit messages, pull request numbers, and author names. This data is untrusted as it can originate from various contributors.
Ingestion points: Git commit logs (SKILL.md Step 2) and pull request metadata (SKILL.md Step 4) are read directly into the agent's context.
Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands within the ingested commit messages.
Capability inventory: The skill possesses significant capabilities including the ability to write to the file system (updating version files and changelogs) and perform network operations via git push and the gh CLI.
Sanitization: The instructions do not specify sanitization or validation of the data parsed from Git logs (e.g., the PR number passed to gh pr view) before it is used in command arguments or changelog generation.
[EXTERNAL_DOWNLOADS]: The skill fetches metadata from GitHub's servers using the gh pr view and gh repo view commands to identify contributors and repository ownership. These are standard operations for a release tool interacting with a well-known service.

release-skills