Probe
AI-friendly semantic code search engine for large codebases. Combines ripgrep speed with tree-sitter AST parsing. Powers AI coding assistants with precise, context-aware code understanding.
What is Probe?
Probe is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to ai-friendly semantic code search engine for large codebases. combines ripgrep speed with tree-sitter ast parsing. powers ai coding assistants with precise, context-aware code understanding.
AI-friendly semantic code search engine for large codebases. Combines ripgrep speed with tree-sitter AST parsing. Powers AI coding assistants with precise, context-aware code understanding.
This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- AI-friendly semantic code search engine for large codebases.
Use Cases
Maintainer
Works with
Installation
NPM
npx -y @probelabs/probeManual Installation
npx -y @probelabs/probeConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Probe
Probe is an AI-optimized semantic code search engine that combines ripgrep's speed with tree-sitter AST parsing to give AI coding assistants precise, context-aware answers about large codebases. It exposes four core MCP tools — Search (boolean keyword queries), Query (structural AST pattern matching), Extract (retrieves complete code blocks by symbol or line), and Symbols (lists all functions, classes, and constants) — and includes token-aware context management so results fit within LLM input budgets. Developers integrate it with Claude Code or other MCP clients to answer architecture questions, perform guided refactoring, and review code across repositories with millions of lines.
Prerequisites
- Node.js (for npm installation) or Rust toolchain (for source builds)
- An MCP client such as Claude Desktop or Claude Code
- Optional: an API key (ANTHROPIC_API_KEY, OPENAI_API_KEY, or GOOGLE_API_KEY) if using Probe's built-in agent mode independently of Claude Code
Install Probe via npm
The npm install is the fastest way to get the Probe binary on any platform with Node.js.
npm install -g @probelabs/probeAlternatively install via curl (macOS/Linux)
A shell script installer is available if you prefer not to use npm.
curl -fsSL https://raw.githubusercontent.com/probelabs/probe/main/install.sh | bashVerify the installation
Confirm the probe binary is accessible and check its version.
probe --versionConfigure Probe as an MCP server in Claude Desktop
Add the Probe server entry to your Claude Desktop configuration file so it is available in every conversation. Probe will search from your working directory by default.
Point Probe at a codebase
When starting a conversation, tell Claude where to search by specifying a directory. Probe will index and search that directory using ripgrep and tree-sitter.
Probe Examples
Client configuration
Claude Desktop configuration for the Probe MCP server using the globally installed npm binary.
{
"mcpServers": {
"probe": {
"command": "npx",
"args": ["-y", "@probelabs/probe"],
"env": {}
}
}
}Prompts to try
Example prompts that leverage Probe's search, query, extract, and symbols tools.
- "Search the codebase for how authentication is implemented and summarize the flow"
- "List all functions in the auth module and show their signatures"
- "Find all usages of the UserService class and identify which methods are called most"
- "Search for 'database NOT sqlite' to find non-SQLite database calls in the project"
- "Extract the full source of the handleLogin function and review it for security issues"Troubleshooting Probe
Search returns no results even for strings that exist in the codebase
Ensure Probe has read access to the target directory and that the path is correct. Probe respects .gitignore by default — if your files are excluded, add them or use the --no-ignore flag.
AST-based Query tool fails on certain file types
Query relies on tree-sitter grammars. Not all languages are supported. Use the Search tool as a fallback for file types that Query cannot parse; it falls back to text matching automatically.
Context window errors when searching very large codebases
Probe's token-aware context management trims results to fit the LLM's input budget. If results are still too large, narrow your search by targeting a specific subdirectory or adding more specific search terms to reduce the match count.
Frequently Asked Questions about Probe
What is Probe?
Probe is a Model Context Protocol (MCP) server that ai-friendly semantic code search engine for large codebases. combines ripgrep speed with tree-sitter ast parsing. powers ai coding assistants with precise, context-aware code understanding. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Probe?
Install via npm with the command: npx -y @probelabs/probe. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with Probe?
Probe works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Probe free to use?
Yes, Probe is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Probe Alternatives — Similar Search & Data Extraction Servers
Looking for alternatives to Probe? Here are other popular search & data extraction servers you can use with Claude, Cursor, and VS Code.
TrendRadar
★ 58.0kA real-time hotspot monitoring and news aggregation assistant that provides AI-powered analysis of trending topics across multiple platforms via the Model Context Protocol. It enables users to track news and receive automated notifications through va
Scrapling
★ 52.7k🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!
PDF Math Translate
★ 33.9k[EMNLP 2025 Demo] PDF scientific paper translation with preserved formats - 基于 AI 完整保留排版的 PDF 文档全文双语翻译，支持 Google/DeepL/Ollama/OpenAI 等服务，提供 CLI/GUI/MCP/Docker/Zotero
GPT Researcher
★ 27.2kAn autonomous agent that conducts deep research on any data using any LLM providers
Agent Reach
★ 20.1kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
Xiaohongshu
★ 13.7kMCP for xiaohongshu.com
Browse More Search & Data Extraction MCP Servers
Explore all search & data extraction servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Probe 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 Probe?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.