Trace
Framework-aware code intelligence MCP server that builds a cross-language dependency graph from source code. 53 integrations (Laravel, Django, Rails, Spring, NestJS, Next.js, and more) across 68 languages. 100+ tools for navigation, impact analysis,
What is Trace?
Trace is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to framework-aware code intelligence mcp server that builds a cross-language dependency graph from source code. 53 integrations (laravel, django, rails, spring, nestjs, next.js, and more) across 68 langu...
Framework-aware code intelligence MCP server that builds a cross-language dependency graph from source code. 53 integrations (Laravel, Django, Rails, Spring, NestJS, Next.js, and more) across 68 languages. 100+ tools for navigation, impact analysis,
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Framework-aware code intelligence MCP server that builds a c
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx traceConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Trace
trace-mcp is a framework-aware code intelligence MCP server that builds a persistent cross-language dependency graph from your source code once, then serves precision context to AI coding agents on demand — replacing repeated file reads and grep chains with single tool calls like get_change_impact, get_task_context, and get_request_flow. It covers 81 framework integrations across 80 languages (Laravel, Django, Rails, Spring, NestJS, Next.js, and more), provides 170 tools for navigation, impact analysis, type hierarchies, test coverage mapping, and cross-session memory, and also indexes Obsidian/Logseq markdown vaults as a peer graph domain — typically reducing agent token consumption by 40-50% on mixed workloads.
Prerequisites
- Node.js (version shown on the npm badge at npmjs.com/package/trace-mcp)
- An MCP-compatible AI coding client: Claude Code, Cursor, Windsurf, Continue, JetBrains AI, or Codex
- A source code repository or markdown vault to index
- Optional: the trace-mcp desktop app (macOS/Windows) for a visual graph explorer
Install trace-mcp via npm
Install the trace-mcp package globally or use npx. The package name on npm is trace-mcp.
npm install -g trace-mcp
# or run without installing:
npx trace-mcpUse the desktop app for one-click MCP client wiring (recommended)
Download the trace-mcp desktop app from GitHub Releases (macOS: trace-mcp-<version>-arm64-mac.zip or Intel zip; Windows: trace-mcp.Setup.<version>.exe). The app detects installed MCP clients (Claude Code, Cursor, Windsurf, Codex, etc.) and wires trace-mcp into them with one click.
# macOS: unzip and drag trace-mcp.app to /Applications
# Windows: run trace-mcp.Setup.<version>.exeIndex your project
Point trace-mcp at your codebase. The daemon listens at http://127.0.0.1:3741 and builds the dependency graph incrementally. Add your project via the desktop app's Projects tab or via the MCP client.
npx trace-mcp --project /path/to/your/projectAdd trace-mcp to your MCP client configuration manually (if not using the desktop app)
If you prefer manual config, add the trace-mcp server entry to your MCP client's config file. The server communicates via stdio.
Start using the 170 code intelligence tools
Once indexed, your AI agent can call tools like get_task_context, get_change_impact, get_request_flow, find_usages, get_untested_symbols, query_decisions, and plan_turn. The graph is updated incrementally as you edit files.
Trace Examples
Client configuration
Manual MCP configuration for Claude Desktop or Cursor pointing at the locally running trace-mcp daemon.
{
"mcpServers": {
"trace-mcp": {
"command": "npx",
"args": ["trace-mcp", "--project", "/absolute/path/to/your/project"]
}
}
}Prompts to try
Example prompts that leverage trace-mcp's graph-powered tools.
- "What would break if I change the User model's email field?"
- "Show me the full request flow from the /api/users route to the rendered page."
- "Which symbols in the auth module have no test coverage?"
- "What architectural decisions were made about the database connection pooling?"
- "Find all callers of the processPayment function across the entire codebase."Troubleshooting Trace
Indexing stalls or the daemon at 127.0.0.1:3741 is unreachable
Check that no other process is using port 3741. Run 'lsof -i :3741' (macOS/Linux) to identify conflicts. Restart the trace-mcp daemon. For large codebases (5000+ files), initial indexing can take several minutes — the desktop app's Overview tab shows live indexing progress.
Framework-aware edges (e.g. Laravel controller to Vue component) are not appearing
Confirm your framework is in trace-mcp's 81 supported integrations (listed in docs/comparisons.md). Some integrations require specific file naming conventions or directory structures. Check the GitHub repository for known issues with your framework version.
The MCP client shows trace-mcp tools but they return empty results
The project must be fully indexed before tools return data. Check the desktop app's Overview tab for indexing status (should show 'Ready'). If indexing fails, look for error messages in the daemon output — common causes are binary files triggering parse errors, which can be excluded via .traceignore.
Frequently Asked Questions about Trace
What is Trace?
Trace is a Model Context Protocol (MCP) server that framework-aware code intelligence mcp server that builds a cross-language dependency graph from source code. 53 integrations (laravel, django, rails, spring, nestjs, next.js, and more) across 68 languages. 100+ tools for navigation, impact analysis, It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Trace?
Follow the installation instructions on the Trace GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Trace?
Trace works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Trace free to use?
Yes, Trace is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
Trace Alternatives — Similar Developer Tools Servers
Looking for alternatives to Trace? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Trace in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Trace?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.