How to Automate Your Job Search in Gemini CLI with JobGPT MCP

Last updated: March 2026

What if you could type “find me remote senior React jobs paying over $150k” into your terminal, and Google’s Gemini actually searched real job boards, saved the matches, and auto-applied for you?

That’s exactly what the JobGPT MCP server does. It connects Gemini CLI to the full JobGPT platform: 34 tools covering job search, applications, resumes, recruiter outreach, and more. No browser tabs. No copy-pasting. Just ask.

What is MCP?

MCP (Model Context Protocol) is an open standard that lets AI assistants connect to external tools and services. Think of it as giving your AI hands. It can reach out, search for jobs, submit applications, and pull back results, all within your conversation.

Google added MCP support to Gemini CLI, letting you connect to remote MCP servers over Streamable HTTP. Once connected, Gemini gains access to the entire JobGPT job search engine and automation platform.

What You Can Do

Once set up, just chat naturally with Gemini. Here are some things you can ask:

• “Search for remote Python developer jobs paying over $150k”
• “Create a job hunt for Senior Engineer roles in San Francisco”
• “Show me my application stats for the last 7 days”
• “Generate a tailored resume for this Google application”
• “Find recruiters for this job and draft an outreach email”
• “Auto-apply to the top 5 matches from my job hunt”

Gemini handles the back-and-forth with JobGPT automatically. You just describe what you want in plain English.

Requirements

• Gemini CLI installed on your machine (install from GitHub)
• Node.js 18+ (check with node --version)
• A JobGPT account with an API key

Setup: Step by Step

Step 1: Get Your API Key

1. Go to 6figr.com/account
2. Scroll to MCP Integrations
3. Click Generate API Key
4. Copy the key (it starts with mcp_)

You’ll need a JobGPT account. If you don’t have one, sign up here. You get 5 free auto-apply credits to start.

Step 2: Add the JobGPT MCP Server

Gemini CLI reads MCP server config from a settings.json file. Open or create the file at:

• macOS/Linux: ~/.gemini/settings.json
• Windows: %USERPROFILE%.geminisettings.json

Add the following configuration. Replace your-api-key-here with the key you copied in Step 1:

{
  "mcpServers": {
    "jobgpt": {
      "httpUrl": "https://mcp.6figr.com/mcp",
      "headers": {
        "Authorization": "Bearer your-api-key-here"
      },
      "timeout": 30000
    }
  }
}

No extra packages to install. Gemini CLI connects directly to the hosted JobGPT server over Streamable HTTP.

If you already have other MCP servers configured, just add the "jobgpt" block inside the existing "mcpServers" object.

Step 3: Start Using It

Launch Gemini CLI in your terminal:

gemini

You should see JobGPT’s tools loaded. Just start asking about jobs, and Gemini will use the right tools automatically.

Alternative: Run the Server Locally

If you prefer to run the JobGPT MCP server on your own machine instead of connecting to the remote server:

{
  "mcpServers": {
    "jobgpt": {
      "command": "npx",
      "args": ["-y", "jobgpt-mcp-server"],
      "env": {
        "JOBGPT_API_KEY": "your-api-key-here",
        "JOBGPT_API_URL": "https://6figr.com"
      }
    }
  }
}

This requires Node.js 18+ installed on your machine.

All 34 Tools at Your Fingertips

The JobGPT MCP server gives Gemini access to 34 tools across six categories:

Category	What You Can Do
Job Search	Search with filters, get job matches, view job details
Profile & Salary	View/update profile, manage compensation, check credits
Job Hunts	Create and manage saved searches with auto-apply
Applications	Track applications, view stats, auto-apply, import jobs by URL
Resume	Upload, generate tailored resumes, calculate match scores
Outreach	Find recruiters/referrers, send personalized outreach emails

Example Workflow

Here’s what a typical job search session looks like with Gemini CLI + JobGPT:

1. You: “Create a job hunt for Senior Frontend Engineer roles in NYC, remote OK, $160k+ minimum”
2. Gemini: Creates the job hunt with your filters and returns the details
3. You: “Find matching jobs”
4. Gemini: Returns fresh job matches you haven’t seen before
5. You: “Add the Stripe and Airbnb jobs to my applications and auto-apply”
6. Gemini: Saves them, triggers auto-apply, confirms submission
7. You: “Generate a tailored resume for the Stripe application”
8. Gemini: Creates an AI-optimized resume highlighting relevant skills
9. You: “Find recruiters at Stripe and draft an outreach email”
10. Gemini: Finds contacts and drafts a personalized message

All of this happens in a single terminal session. No browser tabs. No manual data entry.

---

Troubleshooting

Tools not loading in Gemini CLI?

• Make sure your settings.json is valid JSON (no trailing commas, correct brackets)
• Verify the file is at ~/.gemini/settings.json
• Restart Gemini CLI after editing the config

“API Error (401)” when using tools?

Your API key is invalid or expired. Generate a new one at 6figr.com/account.

“Out of credits” message?

Operations like auto-apply and resume generation consume credits. You can purchase more credits here.

FAQ

Is the JobGPT MCP server free?

The MCP server itself is free and open source. You get 5 free auto-apply credits when you sign up for JobGPT. After that, credits start at $34.99/month.

Does it work outside the US?

Yes. JobGPT supports US, Canada, UK + 150+ countries. The job search covers global listings.

Do I need to install anything besides Gemini CLI?

No. The remote setup connects directly to the hosted JobGPT server. If you choose the local alternative, you’ll need Node.js 18+.

Is Gemini CLI free?

Yes. Gemini CLI is free and open source. You just need a Google account to authenticate.

Can I use this alongside other MCP servers?

Absolutely. Just add JobGPT alongside your other servers in the same settings.json file. They all work independently.

---