Reed Jobs MCP Server for Job Search

Overview

The Reed Jobs MCP Server exposes Reed.co.uk job search and detail endpoints through a Model Context Protocol (MCP) server, letting developers query job listings programmatically from tools that support MCP (for example Cursor IDE). It’s implemented in TypeScript and packaged to be straightforward to run locally or inside an MCP-aware IDE.

This server is useful when you want to integrate Reed’s job data into developer workflows, chat assistants, or local tooling: run filtered searches (keywords, location, contract type, salary ranges), or fetch full details for a specific job ID. It abstracts Reed API authentication and pagination into an easy-to-call MCP toolset so you can focus on building features around search results instead of managing raw API calls.

Features

• Search Reed job listings by keyword and location
• Filter results by contract type (permanent, contract), work pattern (full-time/part-time), and remote options
• Salary range filtering (minimum / maximum)
• Distance-from-location search
• Retrieve full job details by job ID
• Built in TypeScript; optimised for use with Cursor IDE MCP configuration
• Simple environment variable configuration for Reed API key

Installation / Configuration

Prerequisites:

• Node.js v16+
• npm or yarn
• A Reed API key (register at https://www.reed.co.uk/developers)

Install via Smithery (automated install for supported clients):

npx -y @smithery/cli install @kld3v/reed_jobs_mcp --client claude

Manual install:

git clone https://github.com/kld3v/reed_jobs_mcp.git
cd reed_jobs_mcp
npm install

Create a .env file in the project root:

REED_API_KEY=your_reed_api_key_here

Build and run:

npm run build
npm start

Cursor IDE MCP configuration (add to mcp.json):

{
  "mcpServers": {
    "reed-jobs-mcp": {
      "command": "node",
      "args": ["path/to/your/dist/index.js"],
      "cwd": "path/to/your/project"
    }
  }
}

Environment Variables

• REED_API_KEY — (required) your Reed API key string used to authenticate requests to the Reed API.

Available Tools

Two primary MCP tools are exposed by the server:

• mcp_reed_jobs_search_jobs(params)
• mcp_reed_jobs_get_job_details({ jobId })

Function signatures and parameters:

Example TypeScript-style call (conceptual MCP tool invocation):

// Search for remote frontend roles within 10 miles of "London"
mcp_reed_jobs_search_jobs({
  keywords: "frontend developer",
  locationName: "London",
  distanceFromLocation: 10,
  minimumSalary: 40000,
  permanent: true,
  fullTime: true
});

Example get details:

mcp_reed_jobs_get_job_details({ jobId: 1234567 });

Returned objects include standard Reed fields such as jobId, jobTitle, companyName, locationName, salary, contractType, shortDescription, and a URL to the posted job.

Use Cases

• Build a job-search chatbot inside Cursor that can accept natural-language queries (e.g., “Find remote Python jobs in Manchester with salary > 50k”) and return curated results.
• Integrate Reed job feeds into an internal dashboard: schedule periodic searches and display new or modified listings.
• Automate candidate matching: run targeted searches from an applicant tracking system and enrich job records with Reed’s canonical job details by jobId.
• Rapid prototyping: use the MCP server inside a local MCP client or IDE to test UI filters and pagination without writing direct API integration code.

Concrete example: Add a command to your developer tool to surface three highest-salaried permanent jobs for “data engineer” in London:

1. Call mcp_reed_jobs_search_jobs with keywords=“data engineer”, locationName=“London”, permanent=true.
2. Sort results by salary locally (server returns salary fields).
3. Present top three listings with jobTitle, companyName, salary, and apply URL.

Contributing

Contributions are welcome. Typical workflow:

1. Fork the repository
2. Create a feature branch
3. Implement changes and add tests
4. Open a pull request with a clear description of your change

Please follow TypeScript linting and build steps before submitting PRs.

License

This project is distributed under the MIT License. See the repository for the full license text.

Source

Repository and issues: https://github.com/kld3v/reed_jobs_mcp