The Agent Skills Directory

[Code Safety] (SAFE): This skill contains no executable scripts, shell commands, or binary files. It is purely instructional markdown.
[Indirect Prompt Injection] (INFO): While the skill provides instructions for the agent to process external data (commit messages and PR titles), it has no associated capabilities like file system access, network communication, or subprocess execution that could be exploited via injection.
[Prompt Injection] (SAFE): No instructions were found that attempt to bypass the agent's safety filters or override system prompts.

Conventional Commits