Startup Jobs MCP Server
Search live startup jobs — including remote — from Claude, ChatGPT, Cursor, or any AI tool that speaks the Model Context Protocol.
https://api.startup.jobs/mcp
No account needed. Add an API key as a bearer token to attribute your usage and unlock full access when your account has it.
What it's useful for
Typical questions you can ask once the MCP server is connected.
- Job search
- “Find remote senior Rails jobs at startups posted this week.” Filter by role, country, remote, and employment type.
- Company research
- “What is Stripe hiring for right now?” Pull a company profile and all live listings.
- Salary intel
- “Salary percentiles for product designer roles in the US.” Aggregates from listings that disclose pay.
- Hiring trends
- “Show hiring trends for AI engineer roles over the last year.” Posting volume by role and country.
Read-only by design. Agents can search and look up listings — they never post or apply.
Connect your client
Pick your client. Free tier needs no API key; add a bearer token when you want attribution or full access.
Works in Claude on the web, desktop, and mobile.
- Open Settings → Connectors
- Add custom connector
- Paste
https://api.startup.jobs/mcp
Leave authentication empty for free tier. Optionally add a bearer API key for attribution or full access.
Run this in your terminal. The next time you start Claude Code, Startup Jobs will be available.
claude mcp add --transport http startup-jobs https://api.startup.jobs/mcp
Optional API key: add --header "Authorization: Bearer sj_YOUR_KEY".
Add to ~/.cursor/mcp.json (or .cursor/mcp.json in a project), then restart Cursor.
{
"mcpServers": {
"startup-jobs": { "url": "https://api.startup.jobs/mcp" }
}
}
Optional API key:
{
"mcpServers": {
"startup-jobs": {
"url": "https://api.startup.jobs/mcp",
"headers": { "Authorization": "Bearer sj_YOUR_KEY" }
}
}
}
- Open Settings → Apps & Connectors
- Enable developer mode
- Add
https://api.startup.jobs/mcpwith no authentication
Any MCP client that supports Streamable HTTP can connect. Point it at the endpoint — auth is optional for free tier.
- URL
https://api.startup.jobs/mcp- Auth
- None required. Optional
Authorization: Bearer sj_YOUR_KEY(orX-Api-Keyif your gateway reserves Authorization).
Most clients use a JSON config shaped like this:
{
"mcpServers": {
"startup-jobs": { "url": "https://api.startup.jobs/mcp" }
}
}
Exact file paths differ by client (VS Code, Windsurf, Claude Desktop, Zed, Codex, etc.). Check your client's MCP docs for where to paste the config.
Available tools
Eight read-only tools for listings, companies, roles, and market data.
search_jobs
Search live listings with keyword, role, country, remote, and employment type filters
get_job
One job with its full description
get_company
Company profile and live listing count
get_company_jobs
All live listings for one company
list_roles
The role taxonomy, for filtering
list_countries
Countries with jobs, for filtering
job_trends
Job posting volume over time by role and country
salary_benchmarks
Salary percentiles from listings that disclose compensation
Access levels
Start free with no account. Upgrade when you need the full archive.
Free
- All eight tools, no account needed
- Jobs from the last 14 days
- 12-month trend and salary summaries
- 20 requests/minute
Full access
- Full listings archive, no 14-day window
- Structured salary data on every job
- Monthly trend & salary series, up to 3 years
- Company-level salary benchmarks
- 300 requests/minute
Email marc@startup.jobs to get full access.
FAQ
What is a job search MCP server?
An MCP server exposes tools an AI assistant can call. Ours lets Claude, Cursor, or ChatGPT search startup.jobs listings, pull full descriptions, look up companies, and run salary or hiring-trend queries — without you leaving the chat.
Startup jobs or only remote?
The corpus is startup roles. Many are remote; filter with search_jobs when you want remote-only. Hybrid and on-site roles are included too.
How is this different from scraping LinkedIn or Indeed?
You get a hosted endpoint with structured startup listings, company context, and market tools (trends, salary percentiles) — not a scraper you have to run and maintain locally.