The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git and gh (GitHub CLI) to perform repository operations such as checking branch state, pushing code with --force-with-lease, and managing pull requests. These operations are the primary purpose of the skill and are performed within the local project environment.
[SAFE]: A mandatory "Release Plan" approval gate is integrated into the workflow, requiring explicit user consent before any remote mutations or notifications are triggered. Additionally, the skill includes explicit instructions to avoid exposing credentials, tokens, or internal environment data in pull request bodies.
[PROMPT_INJECTION]: The skill processes untrusted repository metadata (branch names, commit logs, diffs) and Jira issue data to generate PR descriptions and Jira updates.
Ingestion points: Data is collected from the repository environment via git log, git diff, and gh pr list as specified in references/github-pr-flow.md.
Boundary markers: Employs a mandatory "Release Plan" in SKILL.md that acts as a human-in-the-loop validation step for all AI-generated content.
Capability inventory: Orchestrates remote mutations including git push and gh pr create/edit.
Sanitization: Instructions require translating external content to Spanish and normalizing titles/bodies into concise, factual formats to reduce the risk of malicious data influencing the agent's behavior.

krt-release-marshal