MCP server for semantic code search and navigation that reduces token waste for AI agents.
sourcerer-mcp is worth checking the docs before setup with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
Repository setup guidance
gh repo view st3v3nmw/sourcerer-mcp --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.
Sourcerer MCP helps AI agents find the right code without reading entire files. It builds a search index of your codebase so agents can jump directly to functions or classes they need. This saves tokens and speeds up development.
Sourcerer MCP is a Model Context Protocol server that enables AI agents to perform semantic code search and navigation. It parses source code using Tree-sitter, extracts meaningful chunks (functions, classes, methods, types), generates embeddings via OpenAI's API, and stores them in a persistent vector database. The server provides MCP tools such as semantic_search, get_chunk_code, find_similar_chunks, index_workspace, and get_index_status. It watches for file changes and automatically re-indexes modified files. Supported languages include Go, JavaScript, Markdown, Python, and TypeScript, with more planned. This approach dramatically reduces token usage by allowing agents to find relevant code without reading entire files.
Strong trust signals; still review the README and permissions before production use.
Last commit was about 259 days ago.
115 GitHub stars indicate community interest.
1 open issues signal maintenance load.
MIT license detected.
AI-powered code navigation in large codebases
Reducing token consumption for LLM-based coding assistants
Semantic search for finding relevant functions or classes
Automated code review and analysis
Integrating with Claude Code or other MCP-compatible agents
Requires OpenAI API key; embeddings are sent to OpenAI servers.
Stores vector database locally in .sourcerer/ directory; ensure it is added to .gitignore.
File system watcher may consume resources on large repositories.
115
Stars
12
Forks
1
Issues
MIT
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.
MCP server for semantic code search and navigation that reduces token waste for AI agents.
Developers using AI coding assistants like Claude Code, Teams working on large codebases needing efficient code search, Open source maintainers looking to improve code discoverability
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
Requires OpenAI API key; embeddings are sent to OpenAI servers. Stores vector database locally in .sourcerer/ directory; ensure it is added to .gitignore.
Similar or complementary options to evaluate include roam-code, code-memory, MCP Server Collection.
Jun 6, 2026
3 security/trust notes recorded.
Setup difficulty is 3/5.