The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the command just full-write if a justfile is detected in the repository to format the changelog and package files.
[COMMAND_EXECUTION]: The skill performs several git operations, including git status, git commit, and git tag, to facilitate the release process.
[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection from processed repository data.
Ingestion points: Data is ingested from package.json, CHANGELOG.md, and git diff outputs.
Boundary markers: The instructions lack delimiters or specific warnings for the agent to ignore instructions embedded in the project's git history or files.
Capability inventory: The skill has the ability to execute shell commands (git, just) and modify local files.
Sanitization: The skill applies a scope filter based on the files field in package.json to limit data ingestion, but it does not perform content-level sanitization on commit messages.

bump-release