The Agent Skills Directory

[SAFE]: The skill is a purely instructional workflow guide defining how to move from a product idea to implementation using documented stages (Idea, Evaluation, Design, PRD). It does not include scripts or binary files.
[SAFE]: The skill utilizes a local docs/ directory for storing project-related markdown files. This is a standard organizational practice for workspace-based AI agents and does not involve access to sensitive system paths or credentials.
[PROMPT_INJECTION]: Analysis of the instructional content shows no attempts to bypass safety filters, override agent behavior, or extract system prompts. The commands described are for internal workflow management.
[PROMPT_INJECTION]: Evaluation of Indirect Prompt Injection surfaces (Category 8): Untrusted data enters the agent context through user-provided idea descriptions. No explicit boundary markers or sanitization logic are defined in the workflow description. However, the workflow relies on standard agent interaction patterns and does not introduce specific vulnerabilities. Ingestion point: User input via the /idea command into file-based storage. Boundary markers: Absent. Capability inventory: Potential code generation and plan execution through downstream commands such as /overnight and /plan. Sanitization: Absent.

product-lifecycle