Get to the protocol 🚁

Does reading docs make your brain hurt? Do you ever wish you could just build a thing?

The SeekrFlow MCP server turns your AI assistant into an expert SeekrFlow developer. Ask natural language questions ( "Where is that dang endpoint?" ) and get working code examples instantly.

Save the reading for your book club*. Start building.

*and if you're in a book club...call me. 📱

What is MCP?

Model Context Protocol (MCP) is an open standard that allows AI applications to securely access external data sources and tools. The SeekrFlow MCP server provides AI agents with:

Direct API access to all SeekrFlow functionality
Intelligent code generation with proper authentication and error handling
Real-time documentation search capabilities
Working examples generated on-demand for any API endpoint

Why use MCP?

Instead of manually reading documentation and writing API integration code, MCP lets you:

Ask natural questions: "How do I upload a file to the AI-Ready Data Engine?"
Get working code instantly: Complete Python/JavaScript examples with authentication
Never leave your editor: Access all SeekrFlow capabilities without switching contexts
Prototype faster: From idea to working code in seconds, not hours

Supported AI development tools

✅ Fully Supported:

Cursor - AI-powered code editor with native MCP integration
Windsurf - Advanced AI coding environment

❌ Not Currently Supported:

Claude Desktop - Claude Desktop only supports local MCP servers 😢

Authentication

You need your SeekrFlow API key - the same one you use for direct API calls. Get it from your SeekrFlow account dashboard and put it directly in the URL. If you don't have one, see Quickstart.

When you're ready to make actual API calls to SeekrFlow, you'll use your SeekrFlow API keys in the generated code - but the MCP server itself requires no authentication.

Quickstart

SeekrFlow hosts a remote MCP server at https://docs.seekr.com/mcp. Configure your AI development tools to connect to this server. If your APIs require authentication, you can pass in headers via query parameters or however headers are configured in your MCP client.

For Cursor

Open Cursor Settings in your CLI using ls -la ~/.cursor/ and navigate to the MCP configuration
Edit ~/.cursor/mcp.json (use touch to create the file if it doesn't exist)
Add SeekrFlow MCP server: Make sure to specify your SeekrFlow API key.

  {
      "mcpServers": {
        "seekr-api": {
          "url": "https://docs.seekr.com/mcp?https://docs.seekr.com/mcp?authorization=Bearer [Your SeekrFlow API key here]"
        }
      }
    }

Restart Cursor to load the MCP connection

For Windsurf

Follow the same configuration format as Cursor, adjusting for Windsurf's MCP configuration location.

Add to~/.codeium/windsurf/mcp_config.json:

  {
      "mcpServers": {
        "seekr-api": {
          "url": "https://docs.seekr.com/mcp?https://docs.seekr.com/mcp?authorization=Bearer [Your SeekrFlow API key here]"
        }
      }
    }

What you can do

Once connected, you can ask your AI assistant natural language questions and get complete, working solutions:

Vector database operations

"How do I create a vector database for my documentation?"
"Show me how to upload PDFs and generate embeddings"
"What's the status of my vector database ingestion job?"

Fine-tuning workflows

"How do I start a fine-tuning job with my training data?"
"Generate context-grounded fine-tuning data for my documents"
"Monitor my model training progress"

Agent creation

"Create a knowledge retrieval agent for customer support"
"How do I configure an agent with custom instructions?"
"Show me how to build a hybrid agent with file search and websearch"

Code generation and integration

"Write Python code to upload files to the AI-Ready Data Engine"
"Generate authentication headers for SeekrFlow API calls"
"Create a complete workflow from document upload to vector search"

Example interaction

You ask: "How do I upload a file into the SeekrFlow AI Ready Data Engine?"

Your AI assistant:

Searches SeekrFlow API documentation automatically
Finds the correct /flow/files endpoint
Generates complete Python code with proper authentication
Provides step-by-step instructions for implementation
Includes error handling and response parsing

Result: Working code in seconds, not hours of documentation reading.

Troubleshooting

Connection issues

Restart editor: MCP connections sometimes need a fresh start after configuration changes
Check your API key: Make sure it's a valid SeekrFlow API key
Verify your URL: Make sure you're using https://docs.seekr.com/mcp?authorization=Bearer [Your SeekrFlow API key here]

File issues

Cursor: Make sure ~/.cursor/mcp.json exists and is valid JSON
Windsurf: Check ~/.codeium/windsurf/mcp_config.jsonexists
JSON syntax: Verify all quotes and brackets are correct

Still not working?

Test your API key with a direct SeekrFlow API call
🙏