How do local embeddings differ from cloud embeddings?
Local embeddings use SentenceTransformers and require no API key; cloud embeddings use LiteLLM and need a provider API key and model identifier.
@cocoindex-io
Basic information
An MCP (Model Context Protocol) server for indexing and querying codebases using CocoIndex. Features semantic code search with natural language queries, incremental indexing for fast updates, multi-language support (Python, JavaScript/TypeScript, Rust, Go, Java, C/C++, C#, SQL, S
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"cocoindex-code": {
"command": "uvx",
"args": [
"--prerelease=explicit",
"--with",
"cocoindex>=1.0.0a13",
"cocoindex-code@latest"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
Fetch the live tool list by running this server in a temporary sandbox using the button above.
Cocoindex Code is an MCP (Model Context Protocol) server for indexing and querying codebases using CocoIndex. It provides semantic code search via natural language queries, supports incremental indexing, multiple programming languages, and flexible embedding models with portable SQLite storage.
Install uv, then add the server via claude mcp add cocoindex-code using uvx. Optionally set COCOINDEX_CODE_EMBEDDING_MODEL and the corresponding API key environment variable. The server exposes a single search tool. Configuration can be adjusted via environment variables like COCOINDEX_CODE_ROOT_PATH.
Local embeddings use SentenceTransformers and require no API key; cloud embeddings use LiteLLM and need a provider API key and model identifier.
The server requires `uv` to run via `uvx`. No external database setup is needed – the index is stored locally in `.cocoindex_code/`.
The index resides in `.cocoindex_code/` under the codebase root, containing a vector database (`target_sqlite.db`) and CocoIndex state (`cocoindex.db/`).
Python, JavaScript, TypeScript, Rust, Go, Java, C, C++, C#, SQL, Shell, Markdown, and Plain Text. Common generated directories are automatically excluded.
If `COCOINDEX_CODE_ROOT_PATH` is not set, the server looks for the nearest parent containing `.cocoindex_code/`, then `.git/`, and finally falls back to the current working directory.


Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.

MCP server for AI Agents — exposes Yocoolab feedback threads and design selections as tools
Set up and manage the Eddyter AI rich text editor by talking to your AI agent — for Claude, Cursor, Codex and other MCP tools. A Model Context Protocol server that lets AI tools add Eddyter to a React, Next.js, Vite or R
MCP server that lets Claude Code, Cursor or Codex ask you for API keys on a local page and write them to .env. The agent only ever sees key names.
Comments