MCP Server

Server URL: https://oceanofjobs.com/mcp

The server uses Streamable HTTP and requires no API key. All tools are read-only and use the same public data as the website.

Connect a clientCopy Anchor Link

Claude CodeCopy Anchor Link

Run this in your terminal:

claude mcp add --transport http oceanofjobs https://oceanofjobs.com/mcp

Start Claude Code and ask it to search for jobs. To check the connection, run /mcp inside a session.

ClaudeCopy Anchor Link

In Claude, open Customize, then Connectors, then Add custom connector. Name it Ocean of Jobs and paste the server URL:

https://oceanofjobs.com/mcp

The server needs no sign-in, so leave the advanced settings empty. Turn the connector on in a chat from the tools menu.

ChatGPTCopy Anchor Link

In ChatGPT, open Settings, then Apps & Connectors. Turn on Developer mode, then choose Create and paste the server URL:

https://oceanofjobs.com/mcp

Choose no authentication. In a new chat, pick the connector from the tools menu.

CursorCopy Anchor Link

Add to Cursor opens Cursor with the server filled in, and you confirm there. To add it by hand, put this in ~/.cursor/mcp.json:

{
  "mcpServers": {
    "oceanofjobs": {
      "url": "https://oceanofjobs.com/mcp"
    }
  }
}

VS CodeCopy Anchor Link

Add to VS Code opens VS Code with the server filled in, and you confirm there. From a terminal instead:

code --add-mcp '{"name":"oceanofjobs","type":"http","url":"https://oceanofjobs.com/mcp"}'

Then start it: run MCP: List Servers from the Command Palette, pick oceanofjobs and choose Start Server. The tools work in Copilot Chat in Agent mode. Check that search_jobs, get_job and get_company are ticked under the tools button.

Gemini CLICopy Anchor Link

gemini mcp add -s user --transport http oceanofjobs https://oceanofjobs.com/mcp

-s user adds it for every folder. Without it, Gemini adds it to the current project only. Gemini loads MCP servers only in folders you have trusted, so gemini mcp list shows the server as disabled until you trust the folder.

CodexCopy Anchor Link

codex mcp add oceanofjobs --url https://oceanofjobs.com/mcp

codex mcp list shows it as enabled.

Hermes AgentCopy Anchor Link

hermes mcp add oceanofjobs --url https://oceanofjobs.com/mcp

Hermes asks whether the server needs authentication. Answer no. It then connects, lists the three tools and asks which to enable. Run hermes mcp test oceanofjobs to check the connection, and start a new session to use the tools.

Kimi CodeCopy Anchor Link

In Kimi Code, run /mcp-config, or edit its mcp.json, and add:

{
  "mcpServers": {
    "oceanofjobs": {
      "url": "https://oceanofjobs.com/mcp"
    }
  }
}

An entry with a url and no transport is an HTTP server. The older kimi-cli is no longer maintained, and its kimi mcp add command no longer saves anything.

Other MCP clientsCopy Anchor Link

For clients that accept a JSON configuration, add the following entry:

{
  "mcpServers": {
    "oceanofjobs": {
      "type": "http",
      "url": "https://oceanofjobs.com/mcp"
    }
  }
}

If your client only supports locally running MCP servers, you can connect through mcp-remote instead:

{
  "mcpServers": {
    "oceanofjobs": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://oceanofjobs.com/mcp"]
    }
  }
}

Available toolsCopy Anchor Link

The server exposes three read-only tools.

search_jobsCopy Anchor Link

Search for open jobs by keyword, company, location, seniority, workplace type, posting age, or minimum salary.

At least one of query, company, city, or country is required. Results are ranked by relevance when a query is provided, and by posting recency otherwise.

Duplicate listings for the same posting are combined into one result. The duplicates_folded field reports how many duplicates were removed.

get_jobCopy Anchor Link

Fetch a job by its ID, returned by search_jobs or get_company.

The response includes the job description as plain text, up to 6,000 characters. Check description_truncated to see whether any text was cut off.

The confidence field indicates where the listing data came from:

• verified - retrieved from the employer's applicant tracking system (ATS) API.
• best_effort - extracted from the employer's careers page.

get_companyCopy Anchor Link

Look up a company by name or domain.

The response includes its current open-job count, jobs added this week, how long it has been tracked, its newest listings, and its Ocean of Jobs page.

If the name matches multiple companies, other_matches lists the alternatives.

Job data and historyCopy Anchor Link

Each job includes two URLs:

• url - the listing on Ocean of Jobs.
• apply_url - the employer's application page.

Job records also include fields that describe how the listing has changed over time.

A reappearance doesn't necessarily mean the employer reposted the job. A temporary failure when checking the employer's board can also make a listing appear to disappear and return. Likewise, an old first_seen date doesn't prove a job was reposted; it means Ocean of Jobs has been tracking that listing for that long.

Limits and cachingCopy Anchor Link

• Each call returns at most 25 results.
• Pagination is limited to an offset of 200. Use the REST API for larger result sets.
• Responses may be cached at the edge for up to five minutes, so repeated calls can return the same data.
• If the server reaches its daily capacity, it returns a JSON-RPC error. Capacity resets at 00:00 UTC.

REST APICopy Anchor Link

You can also access Ocean of Jobs data directly through the public REST API. It returns JSON over HTTPS and requires no API key or account.

Use the REST API when you need larger result sets or more direct control over your requests.