The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard development tools such as git and the GitHub CLI (gh) to manage repository state, stage changes, commit, and push.
[DATA_EXFILTRATION]: The skill includes a specific safety rule to prevent the accidental exposure of sensitive information, instructing the agent to never stage or commit files containing secrets like .env files, credentials, or API keys.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data such as git logs and file contents to determine intent and prepare commits. While this presents a potential surface for indirect prompt injection, the risk is mitigated by the requirement for user confirmation via the AskUserQuestion tool before any commit or push action is finalized.

ship