The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill is designed to ingest untrusted data (user-provided specifications or requirements) and transform it into a structured plan containing executable content.
Ingestion points: Processes user-provided 'spec or requirements' for multi-step tasks (defined in the description).
Boundary markers: Absent. The instructions do not define clear delimiters or include warnings for the agent to ignore instructions embedded within the user's input.
Capability inventory: The skill instructs the agent to write files to the local file system (docs/plans/) and generates plans that include shell commands (git commit, pytest) and Python code intended for execution by other agents or skills.
Sanitization: Absent. There are no instructions to sanitize, validate, or escape content from the user's specification before interpolating it into the final plan document.

writing-plans