godotlens-mcp
An MCP server providing 15 semantic analysis tools for GDScript, powered by Godot's built-in Language Server.
README
GodotLens: AI-First Code Analysis for GDScript
An MCP server providing 15 semantic analysis tools for GDScript, powered by Godot's built-in Language Server.
Built for AI Agents
AI coding agents work with text files but lack semantic understanding of GDScript. When an agent uses grep to find usages of a function, it cannot distinguish a function call from a comment containing the same name, a signal declaration from a signal emission, or an overridden method from an unrelated function.
GodotLens bridges this gap by exposing Godot's built-in Language Server through the Model Context Protocol (MCP), giving AI agents compiler-accurate code intelligence for GDScript — go to definition, find references, diagnostics, rename, and more.
Example: Finding all usages of _on_player_hit:
| Approach | Result |
|---|---|
grep "_on_player_hit" |
12 matches including comments, strings, and similarly named functions |
gdscript_references |
Exactly 4 call sites where _on_player_hit is invoked |
Prerequisites
- Godot 4.x editor must be running with your project open — Godot's LSP server starts automatically when the editor opens a project
- Python 3.10+ (for pip install) or Node.js 16+ (for npx)
Quick Start
Option A: npx (recommended for MCP clients)
Add to your MCP configuration (e.g., .mcp.json for Claude Code):
{
"mcpServers": {
"godotlens": {
"command": "npx",
"args": ["-y", "godotlens-mcp"]
}
}
}
The npm package bundles the full server (~20 KB of Python). Zero external Python dependencies.
Option B: pip
pip install godotlens-mcp
{
"mcpServers": {
"godotlens": {
"command": "godotlens-mcp"
}
}
}
Configuration
| Environment Variable | Default | Description |
|---|---|---|
GODOT_LSP_HOST |
127.0.0.1 |
Godot LSP server host |
GODOT_LSP_PORT |
6005 |
Godot LSP server port |
Tools
Health
| Tool | Description |
|---|---|
gdscript_status |
Check connection to Godot LSP. Use to verify editor is running before other tools. |
Navigation (6 tools)
| Tool | Description |
|---|---|
gdscript_definition |
Navigate to where a symbol is defined. Returns file path and line number. |
gdscript_declaration |
Navigate to the declaration site of a symbol. |
gdscript_references |
Find all references to a symbol across the project. Essential for impact analysis before refactoring. |
gdscript_hover |
Get type information and documentation for a symbol. Use to understand types and return values. |
gdscript_symbols |
List all symbols (classes, functions, variables, signals) in a file. Use to explore file structure. |
gdscript_signature_help |
Get function signature and parameter info at a call site. |
Refactoring
| Tool | Description |
|---|---|
gdscript_rename |
Rename a symbol across all files. Workflow: references to preview impact, rename, then sync. |
Synchronization (3 tools)
| Tool | Description |
|---|---|
gdscript_sync_file |
Sync a modified file with the LSP and get updated diagnostics. Call after editing .gd files. |
gdscript_sync_files |
Batch sync multiple modified files. More efficient than syncing individually. |
gdscript_delete_file |
Notify LSP a file was deleted. Clears stale diagnostics. |
Batch Operations (3 tools)
| Tool | Description |
|---|---|
gdscript_symbols_batch |
Get symbols from multiple files in one call. |
gdscript_definitions_batch |
Get definitions for multiple positions in one call. |
gdscript_references_batch |
Find references for multiple symbols in one call. Use for bulk impact analysis. |
Diagnostics
| Tool | Description |
|---|---|
gdscript_diagnostics |
Get compiler errors and warnings. Workflow: edit, sync, then diagnostics to verify. |
Architecture
┌──────────────┐ ┌────────────────────┐ ┌───────────────────┐
│ AI Agent │ stdio │ GodotLens (MCP) │ TCP │ Godot Editor │
│ (Claude, etc)├────────►│ JSON-RPC 2.0 ├────────►│ Built-in LSP │
│ │◄────────┤ Python 3.10+ │◄────────┤ Port 6005 │
└──────────────┘ └────────────────────┘ └───────────────────┘
GodotLens acts as a bridge between the AI agent and Godot's built-in Language Server. The AI agent communicates with GodotLens via MCP (JSON-RPC over stdio). GodotLens translates MCP tool calls into LSP requests and sends them to the Godot editor over TCP. Responses are compacted for efficient AI consumption.
Zero dependencies — the server uses only the Python standard library. The MCP and LSP protocols are implemented directly, keeping the server lightweight and self-contained.
Important: File Synchronization
Godot's LSP does not automatically detect file changes made outside the editor. When the AI agent modifies a .gd file, it should call gdscript_sync_file or gdscript_sync_files so the LSP re-analyzes the changed code. Without this, diagnostics and navigation results may be stale.
Recommended workflow:
- Use GodotLens tools to analyze code
- Write changes to files
- Call
gdscript_sync_fileto refresh LSP state - Use GodotLens tools to verify changes
Coordinate System
All line and character parameters are 0-indexed, matching the LSP specification:
- Line 0, Character 0 = first character of the file
License
MIT License — see LICENSE for details.
<!-- mcp-name: io.github.pzalutski-pixel/godotlens -->
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.