AI-friendly semantic code search engine that combines ripgrep speed with tree-sitter AST parsing for large codebases.
Probe is easy to set up with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view probelabs/probe --webOpen the official README and confirm the supported install method.
Add the server entry to your MCP client config.
Restart your agent and verify that the server tools appear.
Probe is a tool that helps AI coding assistants understand your codebase. It searches code not just as text, but as structured code, returning complete functions and classes. It works offline, requires no setup, and gives AI agents precise, context-aware results.
Probe is a code and markdown context engine with a built-in agent, designed for enterprise-scale codebases. Unlike traditional grep or embedding-based tools, Probe uses AST-aware structural search to understand code structure. It supports boolean queries (AND, OR, +required, -excluded), BM25 ranking with SIMD acceleration, and token-aware deduplication. Probe can be used as an MCP server, CLI tool, or Node.js SDK. It integrates seamlessly with AI coding assistants like Claude Code, providing complete semantic code blocks in milliseconds. Key features include: zero setup, offline operation, support for 20+ programming languages, and a built-in agent that can search, extract, and reason across your entire codebase. Probe is ideal for spec-driven development, code reviews, onboarding, and any task requiring deep code understanding.
Strong trust signals; still review the README and permissions before production use.
Last commit was about 3 days ago.
628 GitHub stars indicate community interest.
14 open issues signal maintenance load.
Apache-2.0 license detected.
AI coding assistants can use Probe to find relevant code context without embedding models.
Developers can search large codebases for specific functions, classes, or patterns using boolean queries.
Code review tools can leverage Probe to extract complete code blocks for analysis.
Onboarding new team members by quickly finding and understanding code structure.
Spec-driven development where AI agents read and reason about existing code before writing new code.
628
Stars
63
Forks
14
Issues
Apache-2.0
License
Local codebase intelligence CLI and MCP server for AI coding agents with change-safety gates and audit evidence.
An offline MCP server that indexes your codebase for semantic search, code search, and git history retrieval.
Official MCP reference servers from Anthropic. Includes servers for filesystem, GitHub, Postgres, Slack, and more.
0 security/trust notes recorded.
Setup difficulty is 2/5.