The Agent Skills Directory

[SAFE]: The skill implements best practices for software development, including the use of Conventional Commits and Trunk Based Development. All referenced external links are to authoritative documentation sites.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it analyzes Git logs and repository data to generate automated content like commit messages and PR descriptions. 1. Ingestion points: Data entering the agent context includes Git history (log), local file modifications, and platform metadata (Issues/PRs) as referenced in SKILL.md and AGENTS.md. 2. Boundary markers: There are no explicit delimiters or specific instructions for the agent to disregard potential instructions embedded in the Git history or code comments. 3. Capability inventory: The agent is suggested to use MCP tools for Git and GitHub, which have the capability to execute shell commands (git commit/push) and perform API writes. 4. Sanitization: The skill does not provide specific instructions for sanitizing or escaping the data fetched from the repository environment before it is processed by the agent.

git-workflow