MCP server that gives AI assistants accurate React Flow (@xyflow/react) v12 documentation, API references, patterns, and code generation
React Flow MCP demonstrates solid tool design with clear naming conventions (all tools start with action verbs: list, get, search, generate), comprehensive descriptions (avg 120 chars), and well-structured input schemas. All 9 tools have explicit type definitions and parameter descriptions. However, output schemas are not documented in the visible code, tool implementations reference data structures (ApiEntry, patterns, templates) but return types are not formally declared in the tool registration. Error handling is minimal; tools lack recovery guidance and actionable error messages. The server is read-only (no state mutations), which simplifies composition but limits real-world utility. Tool descriptions are adequate but could be more prescriptive about when to use each tool vs. alternatives (e.g., when to call search-docs vs. get-api).
Get a quick reference cheatsheet for React Flow v12 with common imports, patterns, and API signatures
Generate React Flow code from a description. Creates nodes, edges, and component structure based on user input
Get detailed documentation for a specific React Flow API including props, usage, examples, and tips
Get code examples for a specific category (e.g., 'custom-nodes', 'drag-and-drop', 'undo-redo')
Get the complete React Flow v12 migration guide from v11, including breaking changes, type changes, and behavioral changes
Get detailed implementation patterns for common React Flow use cases (e.g., Zustand store, undo/redo, drag-and-drop, auto-layout)
Get ready-to-use code templates for common React Flow components (custom-node, custom-edge, zustand-store)
Output schemas not documented. Tool implementations return complex objects (ApiEntry with props, examples, tips; pattern objects; template code) but the tool registration does not declare return types. LLMs cannot plan downstream operations or extract specific fields without seeing the response structure.
No error handling or recovery guidance. Tools lack descriptions of failure modes (e.g., 'API not found', 'invalid category') and do not provide actionable next steps. An LLM receiving a 404 or validation error has no guidance on how to recover.
Tool descriptions lack prescriptive guidance on when to use each tool. 'search-docs' and 'get-api' both retrieve API information but the descriptions do not explain the distinction (search is keyword-based discovery; get-api is direct lookup). LLMs may waste calls trying both.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 55 | 2026-07-28+ | v2 |
List all available React Flow APIs by kind (component, hook, utility, type)
Search React Flow documentation by keyword. Returns matching APIs and examples ranked by relevance
generate-flow tool lacks constraints on input. The 'description' parameter accepts free-form text with no length limits, format guidance, or examples of valid descriptions. LLMs may pass vague or overly complex descriptions that fail silently.
cheatsheet tool has empty input schema (no parameters). Description is generic ('quick reference cheatsheet'). No guidance on what the cheatsheet contains or when to call it vs. get-api or search-docs.