The Agent Skills Directory

[SAFE]: The skill acts as a set of design and logic instructions for generating a prototype. It leverages standard frontend components and patterns (React conditional rendering, shadcn/ui) that are non-executable in a system-sensitive context.
[INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external business domain data from a DOMAIN.md file, which could theoretically contain instructions intended to influence the agent.
- Ingestion points: The agent is instructed to read DOMAIN.md or .tasks/domain.md to populate UI labels, roles, and business rules.
- Boundary markers: The skill provides specific mapping rules that categorize incoming data into fixed UI sections (Estimate, Schedule, etc.), which naturally constrains how the data is used.
- Capability inventory: The skill is limited to generating UI components (tables, forms, badges) and interacting with browser localStorage. It does not perform network operations or system-level file access.
- Sanitization: No specific sanitization logic is provided in the instructions; the skill relies on the agent's internal safety filters and the static nature of the generated prototype components.

trades-builder