Which integrations are available for this server?

Uses OpenAI's text-embedding-3-small model to create vector embeddings for semantic search across documentation sites and local code repositories.

How do I use Marcus Local MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Marcus Local MCP Server search for 'authentication flow' in my Credo Protocol repo" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Marcus Local MCP Server

by Marcussy34

Overview Schema Related Servers Score Discussions

JavaScript

Local

📚 Marcus Local MCP Server

A Model Context Protocol (MCP) server that indexes documentation sites and local code repositories for semantic search by AI assistants.

Next.js Python MCP ChromaDB

🎯 What Is This?

This is a local MCP server that enables AI assistants (Cursor, Claude Desktop, ChatGPT) to semantically search through:

Documentation websites - Crawled and indexed from any docs site
Local code repositories - All text files from your projects

It uses OpenAI embeddings to create a vector database (ChromaDB) that AI assistants can query through the Model Context Protocol.

Think of it as: Giving your AI assistant instant access to searchable documentation and your entire codebase.

Related MCP server: MCP RAG Server

📋 How It Works

┌─────────────────┐
│  AI Assistant   │ (Cursor, Claude, ChatGPT, etc.)
│  (via MCP)      │
└────────┬────────┘
         │
         ▼
┌─────────────────┐
│   MCP Server    │ (Python - stdio)
│   main.py       │
└────────┬────────┘
         │
         ▼
┌─────────────────┐      ┌──────────────┐
│   ChromaDB      │◄─────┤   OpenAI     │
│  (Vector Store) │      │  Embeddings  │
└────────┬────────┘      └──────────────┘
         │
         ▼
┌──────────────────────┐
│   Indexed Sources    │
│  • Documentation     │
│    - Moca Network    │
│    - Your Docs       │
│  • Repositories      │
│    - Your Codebase   │
│    - Local Projects  │
└──────────────────────┘

The Flow:

Index - Crawl docs OR read local repo files
Chunk - Split content into 800-token chunks
Embed - Create OpenAI embeddings (batched for speed)
Store - Save in ChromaDB vector database
Search - AI assistant queries via MCP protocol
Retrieve - Return relevant chunks from docs/code

🚀 How to Run It

1. Setup

# Clone repository
git clone <your-repo>
cd crawl4ai_test

# Install Node.js dependencies
npm install

# Setup Python virtual environment
python3 -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install Python dependencies
pip install -r mcp-docs-server/requirements.txt

# Install Crawl4AI
pip install -U crawl4ai
crawl4ai-setup

2. Configure

Create .env file in mcp-docs-server/:

OPENAI_API_KEY=your_openai_api_key_here
EMBEDDING_MODEL=text-embedding-3-small
DEFAULT_RESULTS=5

3. Run the Web UI

# Start Next.js server
npm run dev

# Open browser
open http://localhost:3030

4. Connect to Cursor/Claude

Add to your AI assistant config:

For Cursor (~/.cursor/mcp.json or project config):

{
  "mcpServers": {
    "marcus-mcp-server": {
      "command": "/path/to/your/venv/bin/python3",
      "args": ["/path/to/crawl4ai_test/mcp-docs-server/server/main.py"]
    }
  }
}

For Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "marcus-docs": {
      "command": "/path/to/your/venv/bin/python",
      "args": ["/path/to/crawl4ai_test/mcp-docs-server/server/main.py"]
    }
  }
}

📖 How to Use It

Adding Documentation

Via Web UI:

Go to http://localhost:3030
Click "Add New Docs"
Enter:
- URL: https://docs.example.com
- Source Name: Example Docs
- Max Pages: 50 (or unlimited)
Click "Start Indexing"
Wait for completion

Via Command Line:

cd mcp-docs-server
source ../venv/bin/activate

python scripts/crawler.py https://docs.example.com "Example Docs" 50
python scripts/indexer_multi.py "Example Docs"

Adding Repositories

Via Web UI:

Go to http://localhost:3030
Click "Add Repository"
Enter:
- Repository Path: Drag-and-drop folder OR paste path
- Source Name: Auto-generated from folder name
Click "Start Indexing"
Watch live progress

What gets indexed:

✅ All text files (.js, .py, .md, .tsx, .json, .css, etc.)
✅ Auto-skips: node_modules, .git, venv, build, .next, etc.
✅ Batched embeddings (50-100x faster)

Via Command Line:

cd mcp-docs-server
source ../venv/bin/activate

python scripts/repo_indexer.py "/path/to/your/repo" "My Project"

Searching

From Web UI:

Enter query: "How do I initialize the SDK?"
Select source (Docs, Repos, or All)
Click "Search Documentation"
View results

From AI Assistant:

Search all sources:

@marcus-mcp-server search for "authentication flow"

Filter by specific source:

@marcus-mcp-server search for "BorrowInterface component" 
with source="Credo Protocol"

Example usage in Cursor:

User: Using my marcus-mcp-server, show me how authentication 
      is implemented in the Credo Protocol repository

AI: [Searches indexed repository and returns relevant code chunks]

Pro Tip: Always filter by source name to get focused results and save context tokens.

Managing Sources

View Sources:

See all indexed docs and repos on the main page
Filter by "Docs" or "Repos" tabs
Expand to see individual pages/files

Delete Sources:

Click trash icon next to any source
Confirm deletion
Source and all chunks are removed

📁 Project Structure

crawl4ai_test/
├── pages/                        # Next.js UI
│   ├── index.js                 # Main page (search + sources)
│   ├── add.js                   # Add documentation
│   ├── add-repo.js              # Add repository
│   └── api/                     # API routes
│       ├── mcp-search.js        # Search endpoint
│       ├── mcp-info.js          # Get index info
│       ├── add-docs-crawl.js    # Crawl docs
│       ├── add-docs-index.js    # Index docs
│       ├── add-repo-index.js    # Index repository
│       └── mcp-delete-source.js # Delete source
├── components/
│   ├── ui/                      # shadcn/ui components
│   └── home/                    # Page components
├── mcp-docs-server/             # MCP Server
│   ├── server/
│   │   └── main.py             # MCP server (stdio)
│   ├── scripts/
│   │   ├── crawler.py          # Crawl docs with Crawl4AI
│   │   ├── indexer_multi.py    # Index docs
│   │   ├── repo_indexer.py     # Index repositories
│   │   ├── get_source_pages.py # Get pages/files
│   │   ├── search.py           # Search
│   │   └── delete_source.py    # Delete sources
│   ├── data/
│   │   ├── chroma_db/          # Vector database
│   │   ├── chunks/             # Metadata
│   │   └── raw/                # Crawled JSON
│   └── requirements.txt
└── venv/                        # Python environment

🎨 Built With

Frontend: Next.js 15 + shadcn/ui + Tailwind CSS
Backend: Python 3.13 + MCP Protocol
Crawler: Crawl4AI
Vector DB: ChromaDB
Embeddings: OpenAI (text-embedding-3-small)

Status: ✅ Fully Operational | 🤖 MCP Ready | 🔍 Search Enabled

This server cannot be installed

license - not found

quality - not tested

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

–Releases (12mo)

Commit activity

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

Enterprise Code Search MCP
Code Analysis Search Vector Databases
damian-pramparo
A
license
A
quality
D
maintenance
Enables semantic code search across local projects and Git repositories using AI embeddings with ChromaDB. Supports both OpenAI and local Ollama models for private, enterprise-ready code analysis and similar code discovery.
Last updated 2025-09-17
4
4
MIT
MCP RAG Server
RAG Systems Vector Databases Knowledge & Memory
0xrdan
A
license
-
quality
C
maintenance
Enables AI assistants to search and retrieve information from your knowledge base using RAG (Retrieval-Augmented Generation) with hybrid search, document indexing, and ChromaDB vector storage.
Last updated 2026-01-02
34
MIT
RooCode-RAG-Lookup
RAG Systems Vector Databases Search
Mazchoo
F
license
-
quality
D
maintenance
Enables semantic search across documents and code repositories using RAG (Retrieval-Augmented Generation) with vector embeddings. Automatically indexes PDF documents and performs relevance-scored lookups through ChromaDB and sentence transformers.
Last updated 2025-12-21
Claude Context MCP
RAG Systems Vector Databases Code Analysis
sahinrasit
F
license
-
quality
D
maintenance
Enables AI assistants to index and search codebases using semantic search powered by multiple embedding providers (OpenAI, VoyageAI, Gemini, Ollama) and vector database storage.
Last updated 2025-09-26

View all related MCP servers

Related MCP Connectors

Ragora
Search your knowledge bases from any AI assistant using hybrid RAG.
Amber
Long-term memory for AI assistants. Hybrid retrieval, query expansion, auto-topics.
Hint-Services-obsidian-github-mcp
Connect AI assistants to your GitHub-hosted Obsidian vault to seamlessly access, search, and analy…

View all MCP Connectors

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Marcussy34/localMCP-crawl4ai-RAG'

If you have feedback or need assistance with the MCP directory API, please join our Discord server