The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-supplied information (such as venture names, revenue stages, and risk profiles) to generate local files, creating a surface for indirect prompt injection. Since the data is used solely for local documentation and the skill includes prominent legal disclaimers, the risk is minimal.
Ingestion points: Answers to questionnaires SD-1 to SD-11.
Boundary markers: Absent.
Capability inventory: Local file-write operations to ./portfolio/.
Sanitization: Not identified.
[COMMAND_EXECUTION]: The skill performs local filesystem operations, creating directories and files at paths such as ./portfolio/{founder-o-studio-name}/. This functionality uses user-provided strings for directory naming.
[EXTERNAL_DOWNLOADS]: The skill contains references and links to well-known legal and financial resources (e.g., Stripe, Cooley, and Latitud) for informational purposes. These are reputable external services, and the links are presented neutrally.

structure-decision