Hayhooks

v1.0.0Coding Agentsstable

Deploys and serves Haystack Pipelines and Agents as REST APIs and MCP tools, enabling AI assistants to interact with custom NLP pipelines for tasks like RAG, document processing, and intelligent agents through natural language.

aiapiapi-resthaystackllm
Share:
141
Stars
0
Downloads
0
Weekly
0/5

What is Hayhooks?

Hayhooks is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to deploys and serves haystack pipelines and agents as rest apis and mcp tools, enabling ai assistants to interact with custom nlp pipelines for tasks like rag, document processing, and intelligent agent...

Deploys and serves Haystack Pipelines and Agents as REST APIs and MCP tools, enabling AI assistants to interact with custom NLP pipelines for tasks like RAG, document processing, and intelligent agents through natural language.

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Deploys and serves Haystack Pipelines and Agents as REST API

Use Cases

Deploy Haystack pipelines as REST APIs
Enable AI interactions with NLP workflows
Implement RAG and document processing agents
deepset-ai

Maintainer

LicenseApache-2.0
Languagepython
Versionv1.0.0
UpdatedMay 15, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx hayhooks

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Hayhooks

Hayhooks is a Python server that deploys Haystack AI pipelines and agents as both REST APIs and MCP tools, making it possible for AI assistants like Claude to call your custom NLP workflows directly. You define a Haystack pipeline — for RAG, document processing, summarization, or multi-agent orchestration — wrap it in a simple Python file, and Hayhooks automatically generates the MCP tool interface so Claude Desktop or Cursor can invoke it by name. This bridges the gap between production-grade Haystack pipelines and conversational AI assistants without requiring any additional glue code.

Prerequisites

  • Python 3.9 or higher with pip
  • A Haystack pipeline or agent defined using the deepset-ai/haystack library
  • API keys for any LLM providers used in your pipelines (e.g., OPENAI_API_KEY for OpenAI models)
  • An MCP-compatible client such as Claude Desktop or Cursor
  • Basic familiarity with Haystack pipeline concepts
1

Install Hayhooks

Install the hayhooks package from PyPI. Add optional extras if you want the Chainlit chat UI or OpenTelemetry tracing support.

pip install hayhooks
# Optional extras:
# pip install "hayhooks[chainlit]"  # for embedded chat UI
# pip install "hayhooks[tracing]"   # for OpenTelemetry
2

Start the Hayhooks server

Launch the Hayhooks server. It starts on port 1416 by default and immediately begins serving any pipelines you deploy to it.

hayhooks run
3

Deploy a Haystack pipeline as an MCP tool

Deploy your pipeline files to the running Hayhooks server. Each pipeline automatically becomes an MCP tool that AI clients can call by the pipeline name you provide.

hayhooks pipeline deploy-files -n my_rag_pipeline ./my_pipeline_dir
4

Verify the pipeline is available via REST

Test your deployed pipeline through its auto-generated REST endpoint before connecting your MCP client.

curl -X POST http://localhost:1416/my_rag_pipeline/run \
  -H 'Content-Type: application/json' \
  -d '{"question": "What is the company refund policy?"}'
5

Configure your MCP client to connect to Hayhooks

Add Hayhooks to your Claude Desktop or Cursor MCP configuration. The server exposes pipelines as MCP tools over the standard MCP protocol at the same port.

{
  "mcpServers": {
    "hayhooks": {
      "command": "hayhooks",
      "args": ["run", "--port", "1416"]
    }
  }
}

Hayhooks Examples

Client configuration (Claude Desktop)

claude_desktop_config.json entry for connecting Claude Desktop to a locally running Hayhooks server.

{
  "mcpServers": {
    "hayhooks": {
      "command": "hayhooks",
      "args": ["run"],
      "env": {
        "OPENAI_API_KEY": "sk-your-openai-key-here"
      }
    }
  }
}

Prompts to try

Example prompts for using Hayhooks-deployed Haystack pipelines through an AI assistant.

- "Use the my_rag_pipeline tool to answer: what does our employee handbook say about remote work?"
- "Run the document_summarizer pipeline on the uploaded PDF and give me the key points."
- "Call my_agent to search through our product documentation and find anything related to API rate limits."
- "Use the sentiment_pipeline to analyze the tone of these customer reviews."
- "Ask the qa_pipeline: who founded the company and when?"

Troubleshooting Hayhooks

Pipeline deployment fails with import errors

Ensure all Python dependencies for your pipeline components are installed in the same environment as Hayhooks. Run `pip install haystack-ai` and any component-specific packages (e.g., `pip install farm-haystack[inference]`) before deploying.

MCP client cannot connect to Hayhooks server

Verify Hayhooks is running with `curl http://localhost:1416/status`. If you changed the port, update both the Hayhooks start command and any `--port` references. Ensure no firewall is blocking the port.

OpenAI API calls inside pipelines fail

Set the OPENAI_API_KEY environment variable before starting Hayhooks: `export OPENAI_API_KEY=sk-...` and then run `hayhooks run`. Alternatively, pass the env var in your MCP client configuration under the `env` key.

Frequently Asked Questions about Hayhooks

What is Hayhooks?

Hayhooks is a Model Context Protocol (MCP) server that deploys and serves haystack pipelines and agents as rest apis and mcp tools, enabling ai assistants to interact with custom nlp pipelines for tasks like rag, document processing, and intelligent agents through natural language. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Hayhooks?

Follow the installation instructions on the Hayhooks GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Hayhooks?

Hayhooks works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Hayhooks free to use?

Yes, Hayhooks is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "hayhooks": { "command": "npx", "args": ["-y", "hayhooks"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Hayhooks?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides