The Agent Skills Directory

[COMMAND_EXECUTION]: Operating Rule 4 provides instructions for the user to manually install missing skill modules using the npx command. These modules are vendor resources belonging to the same author (wondelai) and are fetched via a well-known service (npm registry).
[DATA_EXPOSURE]: The skill maintains project state by reading and writing to files within a project-specific docs/ directory. This access is limited to project artifacts such as CUSTOMER.md and STRATEGY.md.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests and processes external data provided by the user, such as interview transcripts and customer feedback.
Ingestion points: Data enters the context via docs/CUSTOMER.md, docs/OFFER.md, and direct user input during the intake and validation phases.
Boundary markers: None identified; the instructions do not specify the use of delimiters to separate user-provided data from system instructions.
Capability inventory: The agent is authorized to read and write markdown files within the workspace and to invoke other specialized skills.
Sanitization: There is no evidence of sanitization or validation performed on external content before it is processed or stored in the project documents.

create-business