drawio-mcp-server vs IntelliCode
Side-by-side comparison to help you choose.
| Feature | drawio-mcp-server | IntelliCode |
|---|---|---|
| Type | MCP Server | Extension |
| UnfragileRank | 33/100 | 40/100 |
| Adoption | 0 | 1 |
| Quality | 0 | 0 |
| Ecosystem |
| 0 |
| 0 |
| Match Graph | 0 | 0 |
| Pricing | Free | Free |
| Capabilities | 12 decomposed | 7 decomposed |
| Times Matched | 0 | 0 |
Implements the Model Context Protocol (MCP) specification to expose Draw.io as a callable tool interface for LLM clients like Claude Desktop and oterm. The server receives structured tool calls from MCP clients, translates them into Draw.io operations via a WebSocket-connected browser extension, and returns structured responses back through the MCP protocol. Uses the @modelcontextprotocol/sdk (v1.10.1) for protocol implementation and event-driven message routing through Node.js EventEmitter.
Unique: Uses event-driven architecture with decoupled message bus (bus_request_stream and bus_reply_stream) to separate MCP protocol handling from WebSocket communication, enabling bidirectional LLM-to-Draw.io integration without direct API access
vs alternatives: First MCP server for Draw.io, enabling native integration with Claude and other MCP clients without requiring custom API wrappers or REST middleware
Operates a uWebSockets.js server on port 3000 that maintains persistent WebSocket connections with the Draw.io MCP Browser Extension, enabling real-time bidirectional message exchange. Commands from MCP clients are queued and sent to the extension, which executes them in the Draw.io DOM context and returns results asynchronously. The event bus (Node.js EventEmitter) decouples incoming MCP requests from outgoing WebSocket messages, allowing multiple concurrent diagram operations.
Unique: Uses uWebSockets.js (high-performance C++ WebSocket library) with event-driven message bus decoupling to handle concurrent MCP requests without blocking browser extension communication, enabling non-blocking async operation queuing
vs alternatives: Faster and more responsive than polling-based approaches; event-driven architecture prevents head-of-line blocking when multiple diagram operations are queued simultaneously
Manages WebSocket connection lifecycle with the Draw.io MCP Browser Extension, including initial handshake, connection validation, and graceful disconnection handling. When the extension connects, the server validates the connection, registers event listeners for incoming messages, and begins routing MCP requests to the extension. On disconnection, the server cleans up event listeners and queues pending operations for retry or failure notification to MCP clients.
Unique: Implements explicit handshake validation with the browser extension to ensure protocol compatibility before routing MCP requests, preventing invalid operations on incompatible extension versions
vs alternatives: Handshake validation catches version mismatches early; cleaner than silent failures when extension protocol changes
Maintains a registry of available tools (add-rectangle, update-cell-properties, delete-cell, etc.) with their schemas, descriptions, and input/output specifications. When an MCP client connects, the server exposes this tool registry through the MCP protocol, allowing clients to discover available operations and their parameters. Tools are dynamically loaded from the tool system and registered with their zod schemas, enabling MCP clients to understand tool capabilities without hardcoding.
Unique: Exposes tool registry through MCP protocol with full schema information, enabling LLM clients to understand tool capabilities and constraints without external documentation
vs alternatives: Dynamic tool discovery is more flexible than hardcoded tool lists; schema exposure enables LLM agents to generate valid tool calls without trial-and-error
Provides tools to query the current state of a Draw.io diagram without modifying it: get-selected-cell retrieves properties of the currently selected element, get-shape-categories lists available shape libraries, get-shapes-in-category enumerates shapes within a category, and get-shape-by-name finds specific shapes by name. These tools execute read-only queries through the WebSocket connection to the browser extension, which accesses the Draw.io DOM to extract metadata and return structured JSON responses.
Unique: Implements read-only query tools that execute in the Draw.io DOM context through the browser extension, providing direct access to diagram metadata without requiring diagram export or serialization
vs alternatives: Faster than exporting and parsing diagram XML; provides real-time access to current diagram state without round-tripping through file I/O
Provides tools to create diagram elements (rectangles, circles, diamonds, text, connectors) with validated properties using zod schema validation. Tools like add-rectangle, add-circle, add-diamond, add-text, and add-connector accept structured input parameters (position, size, style, label, connections) that are validated against predefined schemas before being sent to the Draw.io extension. The extension executes the creation in the Draw.io DOM and returns the created element's ID and properties.
Unique: Uses zod schema validation to enforce input correctness before WebSocket transmission, preventing invalid diagram operations from reaching the browser extension and reducing round-trip error handling
vs alternatives: Schema validation at the server layer catches errors early and provides clear error messages to LLM clients; faster than trial-and-error approaches where invalid operations are sent to Draw.io and rejected
Provides tools to modify existing diagram elements after creation: update-cell-properties changes properties of a selected or specified element (label, style, position, size), delete-cell removes elements from the diagram, and style-cell applies predefined or custom styling. Modifications are sent through the WebSocket connection to the browser extension, which updates the Draw.io DOM and returns confirmation with updated element state. Uses event-driven message routing to queue modifications and handle asynchronous responses.
Unique: Separates element creation from modification into distinct tools, allowing LLM agents to create a diagram structure first, then refine properties in a second pass without re-creating elements
vs alternatives: Enables iterative diagram refinement without full diagram regeneration; more efficient than recreating elements when only properties change
Provides the add-connector tool to create connections between diagram elements with validated source and target element IDs. The tool accepts source element ID, target element ID, and optional label/style properties, validates the IDs exist, and sends the connector creation request through WebSocket to the Draw.io extension. The extension creates the connector in the DOM and returns the connector's ID and properties, enabling programmatic relationship mapping in diagrams.
Unique: Validates element IDs before sending connector creation request, preventing orphaned connectors and ensuring diagram structural integrity at the server layer
vs alternatives: Server-side validation prevents invalid connectors from being created in Draw.io; reduces error handling complexity in LLM agents by failing fast with clear error messages
+4 more capabilities
Provides IntelliSense completions ranked by a machine learning model trained on patterns from thousands of open-source repositories. The model learns which completions are most contextually relevant based on code patterns, variable names, and surrounding context, surfacing the most probable next token with a star indicator in the VS Code completion menu. This differs from simple frequency-based ranking by incorporating semantic understanding of code context.
Unique: Uses a neural model trained on open-source repository patterns to rank completions by likelihood rather than simple frequency or alphabetical ordering; the star indicator explicitly surfaces the top recommendation, making it discoverable without scrolling
vs alternatives: Faster than Copilot for single-token completions because it leverages lightweight ranking rather than full generative inference, and more transparent than generic IntelliSense because starred recommendations are explicitly marked
Ingests and learns from patterns across thousands of open-source repositories across Python, TypeScript, JavaScript, and Java to build a statistical model of common code patterns, API usage, and naming conventions. This model is baked into the extension and used to contextualize all completion suggestions. The learning happens offline during model training; the extension itself consumes the pre-trained model without further learning from user code.
Unique: Explicitly trained on thousands of public repositories to extract statistical patterns of idiomatic code; this training is transparent (Microsoft publishes which repos are included) and the model is frozen at extension release time, ensuring reproducibility and auditability
vs alternatives: More transparent than proprietary models because training data sources are disclosed; more focused on pattern matching than Copilot, which generates novel code, making it lighter-weight and faster for completion ranking
IntelliCode scores higher at 40/100 vs drawio-mcp-server at 33/100. drawio-mcp-server leads on quality and ecosystem, while IntelliCode is stronger on adoption.
Need something different?
Search the match graph →© 2026 Unfragile. Stronger through disorder.
Analyzes the immediate code context (variable names, function signatures, imported modules, class scope) to rank completions contextually rather than globally. The model considers what symbols are in scope, what types are expected, and what the surrounding code is doing to adjust the ranking of suggestions. This is implemented by passing a window of surrounding code (typically 50-200 tokens) to the inference model along with the completion request.
Unique: Incorporates local code context (variable names, types, scope) into the ranking model rather than treating each completion request in isolation; this is done by passing a fixed-size context window to the neural model, enabling scope-aware ranking without full semantic analysis
vs alternatives: More accurate than frequency-based ranking because it considers what's in scope; lighter-weight than full type inference because it uses syntactic context and learned patterns rather than building a complete type graph
Integrates ranked completions directly into VS Code's native IntelliSense menu by adding a star (★) indicator next to the top-ranked suggestion. This is implemented as a custom completion item provider that hooks into VS Code's CompletionItemProvider API, allowing IntelliCode to inject its ranked suggestions alongside built-in language server completions. The star is a visual affordance that makes the recommendation discoverable without requiring the user to change their completion workflow.
Unique: Uses VS Code's CompletionItemProvider API to inject ranked suggestions directly into the native IntelliSense menu with a star indicator, avoiding the need for a separate UI panel or modal and keeping the completion workflow unchanged
vs alternatives: More seamless than Copilot's separate suggestion panel because it integrates into the existing IntelliSense menu; more discoverable than silent ranking because the star makes the recommendation explicit
Maintains separate, language-specific neural models trained on repositories in each supported language (Python, TypeScript, JavaScript, Java). Each model is optimized for the syntax, idioms, and common patterns of its language. The extension detects the file language and routes completion requests to the appropriate model. This allows for more accurate recommendations than a single multi-language model because each model learns language-specific patterns.
Unique: Trains and deploys separate neural models per language rather than a single multi-language model, allowing each model to specialize in language-specific syntax, idioms, and conventions; this is more complex to maintain but produces more accurate recommendations than a generalist approach
vs alternatives: More accurate than single-model approaches like Copilot's base model because each language model is optimized for its domain; more maintainable than rule-based systems because patterns are learned rather than hand-coded
Executes the completion ranking model on Microsoft's servers rather than locally on the user's machine. When a completion request is triggered, the extension sends the code context and cursor position to Microsoft's inference service, which runs the model and returns ranked suggestions. This approach allows for larger, more sophisticated models than would be practical to ship with the extension, and enables model updates without requiring users to download new extension versions.
Unique: Offloads model inference to Microsoft's cloud infrastructure rather than running locally, enabling larger models and automatic updates but requiring internet connectivity and accepting privacy tradeoffs of sending code context to external servers
vs alternatives: More sophisticated models than local approaches because server-side inference can use larger, slower models; more convenient than self-hosted solutions because no infrastructure setup is required, but less private than local-only alternatives
Learns and recommends common API and library usage patterns from open-source repositories. When a developer starts typing a method call or API usage, the model ranks suggestions based on how that API is typically used in the training data. For example, if a developer types `requests.get(`, the model will rank common parameters like `url=` and `timeout=` based on frequency in the training corpus. This is implemented by training the model on API call sequences and parameter patterns extracted from the training repositories.
Unique: Extracts and learns API usage patterns (parameter names, method chains, common argument values) from open-source repositories, allowing the model to recommend not just what methods exist but how they are typically used in practice
vs alternatives: More practical than static documentation because it shows real-world usage patterns; more accurate than generic completion because it ranks by actual usage frequency in the training data