dolphin-mcp

A powerful Model Context Protocol (MCP) server that provides Dolphin AI semantic code search capabilities to AI applications like Continue.dev, Claude Desktop, and other MCP-compatible clients.

🚀 Features

• Semantic Code Search: Find code using natural language queries powered by AI embeddings

• Multi-Repository Support: Search across multiple code repositories

• Smart Context Retrieval: Get detailed code chunks and file slices

• VS Code Integration: Generate editor-friendly file URIs

• Robust Error Handling: Graceful fallbacks and informative error messages

• Environment Configuration: Flexible API endpoint configuration

🛠️ Installation

Prerequisites

• Bun (>= 1.0.0): Install Bun

• Dolphin API Server: Running on your specified endpoint

Install the MCP Server

⚙️ Configuration

Environment Variables

Quick Setup

📱 AI Application Integration

Continue.dev Configuration

Add this to your config.yaml:

Claude Desktop Configuration

Add this to your claude_desktop_config.json:

Direct Usage

🧰 Available Tools

The Dolphin MCP server provides these tools to AI applications:

1. search_knowledge

Search codebase semantically using AI embeddings.

Parameters:

• query (string): Search query - Required

• repos (string[]): Optional repository filters

• top_k (number): Number of results (default: 8)

2. fetch_chunk

Get detailed chunk content by ID.

Parameters:

• chunk_id (string): Chunk identifier - Required

3. fetch_lines

Get specific file lines by range.

Parameters:

• repo (string): Repository name - Required

• path (string): File path - Required

• start (number): Start line (1-indexed) - Required

• end (number): End line (inclusive) - Required

4. get_vector_store_info

Get knowledge base statistics and repository info.

Parameters: None

5. open_in_editor

Generate VS Code URI for opening files.

Parameters:

• repo (string): Repository name - Required

• path (string): File path - Required

• start_line (number): Start line (default: 1)

🔧 Development

Prerequisites

• Bun (>= 1.0.0)

• TypeScript

• Node.js 18+

Setup

Project Structure

📋 Requirements

• Bun: >= 1.0.0

• Node.js: 18+

• Dolphin API Server: Accessible via HTTP/HTTPS

🐛 Troubleshooting

Common Issues

Server fails to start:

Connection timeout:

• Verify DOLPHIN_API_URL is correct and accessible

• Check firewall/network settings

• Increase connectionTimeout in your MCP configuration

No search results:

• Ensure repositories are indexed in Dolphin

• Check repository names match exactly

• Verify API credentials and permissions

Logging

The server logs to mcp-bridge/logs/mcp.log with automatic rotation (5MB, 3 files).

🤝 Contributing

1. Fork the repository

2. Create a feature branch (git checkout -b feature/amazing-feature)

3. Commit changes (git commit -m 'Add amazing feature')

4. Push to branch (git push origin feature/amazing-feature)

5. Open a Pull Request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔗 Links

• NPM Package: https://www.npmjs.com/package/@plastic-beach/dolphin-mcp

• GitHub Repository: https://github.com/plasticbeachllc/dolphin-mcp

• MCP Specification: https://modelcontextprotocol.io

• Continue.dev: https://continue.dev

⚡ Quick Start Commands