Add semantic code search to Claude Code and AI agents using vector databases.
Make your entire codebase the context for any coding agent using semantic search.
Claude Context is an MCP plugin that indexes your codebase into a vector database for semantic search. It allows coding agents like Claude Code to retrieve relevant snippets from millions of lines without loading entire directories.
It integrates via the Model Context Protocol to provide semantic code search across large repositories. Instead of multi-round discovery or expensive full-directory context loads, it brings related code straight into your AI agent.
It suits developers working with massive codebases using AI coding assistants like Claude Code, Cursor, or Cline. It does not suit teams wanting a zero-config local setup since it requires external vector database and embedding services.
Finds relevant code across millions of lines using vector embeddings.
Integrates natively with Claude Code, Cursor, Cline, Windsurf, and other MCP clients.
Only sends related code into the agent's context to reduce token costs on large codebases.
Leverages Zilliz Cloud and OpenAI embeddings to store and query your codebase.
The software is free under the MIT license, but running it requires third-party API services including Zilliz Cloud and OpenAI which have their own usage pricing.
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
It supports Claude Code, Cursor, Codex CLI, Gemini CLI, Qwen Code, Void, Claude Desktop, Windsurf, VS Code, Cherry Studio, and Cline.
You need a Zilliz Cloud account for vector storage and an OpenAI API key for generating code embeddings.
See Zekai first in Google