The Agent Skills Directory

[SAFE]: The skill is a meta-prompt designed to guide the AI in creating detailed implementation plans. It defines templates for task decomposition, including documentation of files to be modified and commands to be run.
[COMMAND_EXECUTION]: While the skill instructions mention shell commands like git and pytest, these are provided as examples of what should be included in the generated plan for a human or another agent to follow. The skill itself does not execute these commands autonomously.
[DATA_EXFILTRATION]: The skill instructs the agent to save plan files to a specific directory (docs/superpowers/plans/) within the workspace. No network operations or access to sensitive global files (like SSH keys or environment variables) are present.
[PROMPT_INJECTION]: The skill does not contain instructions to bypass safety filters or override agent behavior. It functions as a set of developer-focused guidelines.
[REMOTE_CODE_EXECUTION]: There are no patterns suggesting the download or execution of remote scripts or unverified packages.

writing-plans