The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references a repository for additional components (pedroromeroluna/ai-first-product-skills) which is handled via a skill installation command. This is identified as a vendor-owned resource belonging to the author.
[COMMAND_EXECUTION]: Instructs the agent to provide the user with shell commands (npx skills add ...) to install supplementary skills required for the full product suite.
[INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted data from preceding agents or skills in the workflow.
Ingestion points: Reads project-specific data and prototype briefs from .agents/skills/ideation-engine/SKILL.md and .agents/skills/prd/SKILL.md.
Boundary markers: Enforcement of a rigid document structure with five literal headings and a specific methodology (Method) that limits the agent's creative freedom.
Capability inventory: Performs file-write operations to the local research/ directory and facilitates interaction with external design tools via the Figma MCP.
Sanitization: Includes a core instruction ("Rule number one") forbidding the agent from filling in missing data with plausible inventions, requiring gaps to be documented explicitly.

visual-designer