The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external specifications which represents a potential surface for indirect prompt injection.
Ingestion points: User-provided specifications or requirements in SKILL.md.
Boundary markers: The instructions do not define specific delimiters for separating the spec content from the instructions.
Capability inventory: The skill saves files to the local filesystem and generates shell commands for testing and version control.
Sanitization: No explicit sanitization of input specifications is performed.
[COMMAND_EXECUTION]: The skill instructs the agent to generate plans containing shell commands, specifically for running tests with pytest and managing commits with git.
[SAFE]: The skill's operations, including saving documents to docs/superpowers/plans/ and referencing other development-focused skills, are consistent with its primary purpose as a technical planning tool.

writing-plans