The Agent Skills Directory

[SAFE]: The skill defines a structured methodology for requirement gathering and design documentation, emphasizing human-in-the-loop approval before implementation begins.
[COMMAND_EXECUTION]: The skill instructs the agent to perform git commits for design documentation. This is a standard project management activity and is limited to recording the design process within the local repository context.
[INDIRECT_PROMPT_INJECTION]: The skill reads project files and documentation to understand context. While this is an ingestion surface for potentially untrusted data, the mandatory requirement for the agent to present designs for human approval before any implementation occurs acts as a significant safety boundary.

brainstorming