MCP server for @florexlabs/ui — lets AI assistants understand and generate code using Florex UI components
Florex UI MCP has clear naming (verb_noun pattern) and basic descriptions for all 5 tools. However, input schemas are minimal or empty for 3 of 5 tools, parameter descriptions are sparse, and output schemas are undocumented. Error handling is present but generic. The server reads component metadata from a registry and generates code, but lacks the structured output documentation and parameter constraints needed for reliable LLM composition.
Generate TSX code using @florexlabs/ui components for a given use case description
Get documentation, props, and usage examples for a specific @florexlabs/ui component
Get the Florex Labs design tokens (colors, spacing, radius, shadows, fonts)
Get installation and setup instructions for @florexlabs/ui
List all available @florexlabs/ui components organized by category
Three tools (list_components, get_design_tokens, get_setup_instructions) have empty input schemas ({}).
Output schemas are not documented. Callers cannot predict response structure (e.g., does get_component_docs return props as array or object? What fields does each prop have?). LLMs must infer structure from examples.
get_component_docs accepts a 'component' parameter but does not constrain it to valid component names via enum. The tool performs a case-insensitive lookup and returns an error if not found, but LLMs cannot discover valid options without calling list_components first.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 49 | 2026-07-28+ | v2 |
generate_component_code uses keyword matching to select components. The algorithm is opaque to the LLM, it cannot predict which components will be included for a given description. No documentation of the matching strategy or fallback behavior.
Error responses (e.g., 'Component not found') lack recovery guidance. Per pattern:recovery-guide, errors should suggest next steps: 'Component not found. Call list_components to see available options.'