The Agent Skills Directory

[SAFE]: The skill provides structured instructions for interactive Q&A and documentation. It manages local markdown files in a predictable manner and uses standard tool invocations within the vendor's ecosystem.- [INDIRECT_PROMPT_INJECTION]: The skill defines a process where user-provided answers are ingested and written into documentation that serves as context for downstream tools like research and planning skills. This creates a standard surface for indirect prompt injection common to collaborative drafting tools.
Ingestion points: User responses provided via AskUserQuestion during the Step 3 loop are appended to markdown files in the thoughts/ directory.
Boundary markers: The skill uses markdown headers (### Q:, [User's answer]) to delimit content, providing basic structural separation.
Capability inventory: The skill utilizes file system writes to create and update brainstorm documents, and it facilitates the invocation of sub-skills (/learning, /research, /create-plan, /file-review) that process the resulting document.
Sanitization: The skill does not explicitly describe sanitization or validation of the user's input before it is written to the shared documentation.

brainstorming