Convert source code repositories into searchable knowledge bases with MCP support for Claude and Cursor.
src-to-kb is easy to set up with trust notes worth reviewing. Check agent compatibility and use-case fit before adding it to your workflow.
Repository setup guidance
gh repo view vezlo/src-to-kb --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.
This tool scans your codebase, splits it into smart chunks, and creates a searchable knowledge base. It can generate embeddings for semantic search and integrates with AI coding assistants like Claude Code and Cursor via the Model Context Protocol. You can also run a REST API server to query your codebase remotely.
src-to-kb is a command-line tool that transforms any source code repository into a structured, searchable knowledge base. It supports multiple programming languages (JavaScript, TypeScript, Python, Java, C++, Go, Rust, etc.) and offers intelligent chunking with configurable overlap. The tool can optionally generate OpenAI embeddings for semantic search, clean code by removing comments and normalizing whitespace, and produce comprehensive statistics about your codebase. It includes several commands: src-to-kb for generating the knowledge base, src-to-kb-search for searching, src-to-kb-upload for uploading to an external server, src-to-kb-api for starting a REST API with Swagger documentation, and src-to-kb-mcp for running an MCP server that integrates with Claude Code and Cursor. The MCP server allows AI assistants to directly query your codebase. The tool also supports Notion integration for importing pages and databases, and can send content to external servers for processing. It features three answer modes (End User, Developer, Copilot) and optional API key authentication for secure access.
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 209 days ago.
33 GitHub stars indicate community interest.
1 open issues signal maintenance load.
NOASSERTION license detected.
Create a searchable knowledge base from a large legacy codebase to help new developers onboard faster.
Integrate with Claude Code or Cursor to enable AI-assisted code search and understanding during development.
Generate embeddings and run a REST API to power a custom code search engine for your team.
Import Notion documentation alongside source code to create a unified knowledge base.
Automatically chunk and index code from multiple repositories for centralized code analysis.
The tool sends code to OpenAI if embeddings are enabled; ensure compliance with data privacy policies.
Running the MCP server exposes your codebase to AI assistants; restrict access via API keys if needed.
External server integration sends code to a remote server; verify the server's security and privacy practices.
33
Stars
8
Forks
1
Issues
NOASSERTION
License
Local code intelligence MCP server and CLI for AI coding agents, providing semantic search and call graph analysis.
A project template that turns your repository into stable infrastructure for AI-assisted development.
Up-to-date code documentation for LLMs and AI code editors, eliminating outdated or hallucinated API references.
Convert source code repositories into searchable knowledge bases with MCP support for Claude and Cursor.
Software developers who want to make their codebase searchable and AI-accessible., DevOps engineers integrating code knowledge into CI/CD pipelines or internal tools., Technical leads onboarding new team members to large projects.
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
The tool sends code to OpenAI if embeddings are enabled; ensure compliance with data privacy policies. Running the MCP server exposes your codebase to AI assistants; restrict access via API keys if needed.
Similar or complementary options to evaluate include codanna, attractor-guided-engineering-template, context7.
Jun 7, 2026
3 security/trust notes recorded.
Setup difficulty is 2/5.