The CodeAtlas MCP Server exposes codebase analysis data to AI assistants, enabling them to query project structure, dependencies, and AI-generated insights.
List Projects (list_projects): Discover all CodeAtlas-analyzed projects under your home directory, returning their names, paths, and last analysis timestamps.
Get Project Structure (get_project_structure): Retrieve all modules, classes, functions, and variables in a project, with optional filtering by entity type and result limits.
Get Dependencies (get_dependencies): Explore import, call, and containment relationships between entities, with filtering by source, target, and relationship type.
Get Insights (get_insights): Access AI-generated code quality insights including refactoring suggestions, security issues, and maintainability analysis.
Search Entities (search_entities): Fuzzy search for functions, classes, modules, or variables by name across a project, with optional type filtering.
Get File Entities (get_file_entities): Retrieve all classes, functions, and variables defined within a specific file using partial path matching.
Compatible with major AI tools including Gemini, Claude, Cursor, Windsurf, and VS Code Copilot. Supports both npx execution and global installation.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@CodeAtlas MCP Server show the module structure and dependencies for the current project"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Give your AI assistant deep understanding of any codebase
A standalone MCP server that exposes code analysis data to Gemini, Claude, Cursor, Windsurf, VS Code Copilot — with persistent AI memory between conversations.
Your Code → CodeAtlas Analyze → MCP Server → AI Understands Everything
│
10 powerful tools:
• Search functions & classes
• Trace feature flows
• Generate architecture diagrams
• Persistent memory across conversationsBefore CodeAtlas: AI greps blindly, forgets your project every conversation. After CodeAtlas: AI instantly knows your architecture, dependencies, and remembers context.
Related MCP server: Codebase MCP Server
Install the CodeAtlas extension, then:
Ctrl+Shift+P → CodeAtlas: Analyze Project{
"mcpServers": {
"codeatlas": {
"command": "npx",
"args": ["-y", "@giauphan/codeatlas-mcp"]
}
}
}{
"mcpServers": {
"codeatlas": {
"command": "npx",
"args": ["-y", "@giauphan/codeatlas-mcp"]
}
}
}{
"mcp": {
"servers": {
"codeatlas": {
"command": "npx",
"args": ["-y", "@giauphan/codeatlas-mcp"]
}
}
}
}{
"mcpServers": {
"codeatlas": {
"command": "npx",
"args": ["-y", "@giauphan/codeatlas-mcp"]
}
}
}claude mcp add codeatlas -- npx -y @giauphan/codeatlas-mcp{
"mcpServers": {
"codeatlas": {
"command": "npx",
"args": ["-y", "@giauphan/codeatlas-mcp"]
}
}
}Done! Your AI can now use all 10 CodeAtlas tools.
Tool | Description | Example Use |
| List all analyzed projects | "What projects do I have?" |
| Get modules, classes, functions | "Show me all classes" |
| Import/call/containment relationships | "What does UserService depend on?" |
| Code quality & security analysis | "Any security issues?" |
| Fuzzy search entities by name | "Find the login function" |
| All entities in a specific file | "What's in auth.ts?" |
Tool | Description | Example Use |
| Mermaid architecture diagram (module imports) | "Show me the system architecture" |
| Mermaid execution flow diagram (call chains) | "How does the payment feature work?" |
Tool | Description | Example Use |
| Save context to | "Remember what we changed" |
| Trace feature through codebase | "What files are involved in auth?" |
AI assistants forget everything between conversations. CodeAtlas fixes this:
Conversation 1 → AI analyzes code → sync_system_memory
│
.agents/memory/
├── system-map.md ← Architecture diagram
├── modules.json ← All entities
├── conventions.md ← Code patterns
├── business-rules.json ← Domain logic
├── feature-flows.json ← Feature traces
└── change-log.json ← Change history
│
Conversation 2 → AI reads memory → full context restored instantly ✨When you run Analyze Project, CodeAtlas auto-creates rule files for your AI IDE:
Generated File | For |
| All AI assistants |
| Cursor |
| Claude Code |
| Windsurf |
These tell your AI to:
Read .agents/memory/ at the start of every conversation
Use MCP tools before making changes
Call sync_system_memory after completing changes
Language | Features |
TypeScript / JavaScript | Full AST: imports, classes, functions, variables, calls, implements |
Python | Classes, functions, variables, imports, calls |
PHP | Classes, interfaces, traits, enums, functions, properties |
Blade Templates |
|
npm install -g @giauphan/codeatlas-mcpThen use "command": "codeatlas-mcp" in your MCP config (no args needed).
Variable | Description |
| Force a specific project directory |
By default, the server auto-discovers all projects with
.codeatlas/analysis.jsonunder~/.
git clone https://github.com/giauphan/codeatlas-mcp.git
cd codeatlas-mcp
npm install
npm run build
npm test # 6 tests
npm start # Start MCP serverCodeAtlas Extension — VS Code extension with interactive code graph
MCP Protocol — Model Context Protocol standard
MIT — Free for personal and commercial use.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Give your AI agent a persistent map of your project's structure, dependencies, and bugs.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Securely search and manage workspace context files for AI agents and teams.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/giauphan/codeatlas-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server