Skip to main content
Glama
wjlim
by wjlim
README.md
# nf-core MCP Server

An MCP server for managing and navigating nf-core pipeline repositories.

## Features

- List local nf-core repositories (rnaseq, sarek, modules, tools)
- Access pipeline configurations and workflows
- Search through pipeline files
- Explore pipeline modules

## Installation

### NPM Version

```bash

cd nf-core_mcp
npm install

# Build TypeScript
npm run build

# Start the server
npm start
```

### Docker Version

```bash
# Build the Docker image
cd nf-core_mcp
docker build -t nf-core-mcp .
# Run the container
docker run -i --rm \
  -v "/path/to/your/workspace:/app/workspace" \
  nf-core-mcp
```

## Adding nf-core Repositories

To add new nf-core pipeline repositories to the workspace:

1. **Clone the repositories**:
   ```bash
   # Navigate to your workspace directory (example for Windows)
   cd /path/to/your/workspace

   # Clone desired nf-core repositories
   git clone https://github.com/nf-core/rnaseq.git
   git clone https://github.com/nf-core/sarek.git
   git clone https://github.com/nf-core/modules.git
   # Add any other nf-core pipeline you want to manage
   ```

2. **Directory Structure**:
   Your workspace should look like this:
   ```
   workspace/
   ├── rnaseq/
   ├── sarek/
   ├── modules/
   └── your-new-pipeline/
   ```

3. **Verify Installation**:
   After starting the MCP server, use the `list-pipelines` command to verify that your new pipelines are detected:
   ```
   list-pipelines
   ```

Note: The MCP server will automatically detect and manage any nf-core pipeline repositories in your workspace directory.

## Available Tools

1. `list-pipelines`
   - Lists all nf-core pipelines in the workspace
   - Shows configuration file status
   - No parameters required

2. `get-pipeline-modules`
   - Gets module information from a pipeline
   - Parameters:
     - `pipeline`: Pipeline name (rnaseq, sarek, or modules)

3. `search-pipelines`
   - Searches through pipeline files
   - Parameters:
     - `query`: Search query
     - `pipeline` (optional): Specific pipeline to search

## Available Resources

1. `pipeline-config`
   - Gets pipeline configuration
   - URI format: `pipeline://{name}/config`
   - Parameters:
     - `name`: Pipeline name (rnaseq, sarek, or modules)

2. `pipeline-workflow`
   - Gets pipeline workflow
   - URI format: `pipeline://{name}/workflow`
   - Parameters:
     - `name`: Pipeline name (rnaseq, sarek, or modules)

## Usage with Cursor IDE

### Using NPX (Recommended)

Add the following to your `mcp.json`:

```json
{
  "mcpServers": {
    "nf-core": {
      "command": "npx",
      "args": ["-y", "nf-core-mcp"]
    }
  }
}
```

### Using Docker

Add the following to your `mcp.json`:

```json
{
  "mcpServers": {
    "nf-core": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-v",
        "/path/to/your/workspace:/app/workspace",
        "nf-core-mcp"
      ]
    }
  }
}
```

## Example Usage

Using the MCP server in Cursor:

```
# List available pipelines
list-pipelines

# Get modules from rnaseq pipeline
get-pipeline-modules pipeline=rnaseq

# Search in all pipelines
search-pipelines query="fastqc"

# Search in specific pipeline
search-pipelines query="fastqc" pipeline=rnaseq

# Access pipeline configuration
pipeline://rnaseq/config

# Access workflow
pipeline://rnaseq/workflow
```

## Running the Server

### Using NPM

```bash
# If installed globally
nf-core-mcp

# If installed locally
npx nf-core-mcp

# Using npx without installation
npx -y nf-core-mcp
```

### Using Docker

```bash
docker run -it --rm \
  -v /path/to/your/workspace:/app/workspace \
  nf-core-mcp
```

## Development

```bash
# Install dependencies
npm install

# Build TypeScript
npm run build

# Run in development mode
npm run dev

# Run tests
npm test

# Run linter
npm run lint
```

## License

MIT

TDQS

D1.8/5.0

Scored across 3 tools

Disambiguation5/5

The three tools have clearly distinct purposes: get-pipeline-modules retrieves modules for a specific pipeline, list-pipelines enumerates available pipelines, and search-pipelines finds pipelines based on criteria. There is no overlap in functionality, making tool selection straightforward for an agent.

Naming Consistency5/5

All tool names follow a consistent verb-noun pattern with hyphens (e.g., get-pipeline-modules, list-pipelines, search-pipelines). The naming is uniform and predictable, using clear action verbs paired with relevant nouns.

Tool Count3/5

With only three tools, the server feels thin for a pipeline management domain, potentially lacking operations like create, update, or delete pipelines. However, the tools cover basic discovery and inspection, which might be sufficient for a limited scope.

Completeness2/5

The toolset is severely incomplete for pipeline management, missing essential CRUD operations such as creating, updating, or deleting pipelines. It only supports discovery and inspection, leaving significant gaps that could hinder agent workflows.

Maintenance

ActivityInactive
ResponsivenessNo issues