M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
M

MCP Jobs

Community
Zero-config multi-platform job aggregation MCP server
How to install ↓ Compare GitHub source repository ↗
Category
Other #224 of 230
Stars
★ 139 Popular
Transport
stdio (local process)
Runtime
Node.js 16+
Credentials
No credential needed
Last commit
⚠
Tools
2
26FMRS · D

This project offers zero-config job search functionality, suitable for individuals and developers to quickly integrate into AI assistants, but relies on scraping, posing compliance and stability risks.

Strongest · Maintenance 7/20 Weakest · Security and permissions 3/20

Reliability
4/20
Security and permissions
3/20
Maintenance
7/20
Documentation
6/20
Setup experience
6/20
Why each score
Reliability 4/20
Source code files were not provided, so we cannot verify whether the server starts cleanly, completes the MCP init handshake, or whether declared tools match real behavior. The use of web scraping to fetch data from third-party job sites introduces potential instability due to site changes or anti-bot measures. Error handling and dependency control are not assessable. Therefore, a low score is warranted.
Security and permissions 3/20
Source code is unavailable, so credential handling, least privilege, data boundaries, external network requests, and side effects cannot be evaluated. The server claims to work without API keys but likely relies on scraping job sites, which raises potential legal and ethical concerns. Without source evidence, we cannot confirm data-flow disclosure or permission scoping. No explicit red-line violation is found, but the overall security posture is unclear, hence a low score.
Maintenance 7/20
The GitHub repo shows 114 stars and 3 open issues, and it is not archived, indicating some activity. However, there is no evidence of releases, commit history, issue response times, or dependency update policies. The README mentions an MIT license, but we cannot confirm a LICENSE file exists in the repository. Given this lack of governance evidence, the score is limited.
Documentation 6/20
The README is comprehensive, providing installation, quick start, client configuration examples, and optional environment variables. However, it lacks detailed API documentation (e.g., tool parameters, response formats), troubleshooting guidance, and disclosure of limitations (e.g., rate limits, anti-scraping measures). Thus, documentation is usable but has critical gaps.
Setup experience 6/20
The README clearly describes a zero-configuration installation via npx and provides concrete configuration examples for multiple clients, which is a strong setup story. However, the source code is not provided, so we cannot verify that the package is actually available or that the instructions work. Additionally, potential dependencies (e.g., Node version, browser dependencies) are not detailed. Without execution evidence, the score remains moderate.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkControls a browser

Best for

  • Users who need quick access to job listings from major Chinese recruitment sites
  • Developers who want to integrate natural language job search into AI assistants
  • Those who prefer a zero-configuration, out-of-the-box setup

Not for

  • Scenarios requiring official API support or enterprise-grade reliability
  • Compliance-sensitive use that must strictly adhere to recruitment websites' terms of service (since it uses scraping)
  • Users who need global job coverage

Required permissions

  • No extra permissions required, but the service accesses the internet to scrape recruitment websites

Risks and side effects

  • Dependence on scraping may violate target websites' terms and risk blocking
  • Data timeliness and accuracy may be affected by anti-scraping measures
  • No official API, so functionality may break if website structures change

Setup

Before you start

Runtime：Node.js 16+

USERNAME optionalsecret Optional basic-auth username you set yourself in a .env file; no third-party signup needed.
PASSWORD optionalsecret Optional basic-auth password paired with USERNAME, set in a .env file.
Other optional settings (6)
CRAWLER_HEADLESS optional Optional; controls headless browser mode, set to false for debugging.
CRAWLER_TIMEOUT optional Optional; page load timeout in milliseconds, default around 30000.
CRAWLER_VIEWPORT_WIDTH optional Optional; browser viewport width, e.g. 1280 or 1920.
CRAWLER_VIEWPORT_HEIGHT optional Optional; browser viewport height, e.g. 800 or 1080.
CRAWLER_DEBUG optional Optional; set to true to enable debug logging.
CRAWLER_USER_AGENT optional Optional; custom crawler User-Agent string.

Run npx -y mcp-jobs in a terminal to start the service, then add an MCP server in your AI client (e.g., Claude Desktop, Cursor) with command npx -y mcp-jobs and empty environment variables. No registration or API key needed.

claude_desktop_config.json 
{
  "mcpServers": {
    "mcp-jobs": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-jobs"
      ]
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "mcp-jobs": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-jobs"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add mcp-jobs -- npx -y mcp-jobs

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After adding the server, search_jobs and get_job_details should appear in the client's tool list; asking "Search for frontend developer jobs in Beijing" and getting job results confirms the connection works.

Troubleshooting

  1. Check that Node.js version is >= 16.0.0; ensure network connectivity; if scraping issues occur, try setting CRAWLER_HEADLESS=false for debugging or adjust CRAWLER_TIMEOUT and other environment variables.

Things to try

Once connected, you can ask your AI assistant things like:

  • Search for frontend developer jobs in Beijing
  • Find data analyst roles in Shanghai requiring 3+ years of experience
  • Get the details of this job: [job link]
  • Compare salaries for the same role in Hangzhou and Beijing

Tools 2

search_jobs read-only
Search for jobs by title, city, salary range, etc.
get_job_details read-only
Get detailed information for a specific job, including requirements and salary.

Use cases

Quickly search for job listings in a specific city and role
Fetch job details to analyze requirements and compensation
Compare salary levels for similar roles across cities

Supported clients

Claude Desktop
Cursor
Windsurf
Cline

Listed from the project's documentation, not tested by this site.

Overview

mcp-jobs is an MCP-based job aggregation service that fetches job listings from major Chinese recruitment websites like Liepin, Boss, Zhilian, and 51job. It offers a zero-configuration experience with no API keys or registration required. Start with a single command, search jobs, get details, and aggregate data from multiple platforms, with standardized output for easy AI processing.

Similar servers

BioMCP 75 · B

One binary. One grammar. Evidence from the biomedical sources you already trust.

★ 654 · Tools 11 Compare with this →
Compare with other MCP servers

Source revision 23e223d34fc1 Data synced 2026-10-12 Read the FMRS scoring method