openings-mcp searches job listings across job boards and company career sites from any MCP client: Claude Code, Codex, Gemini CLI, and others.

⚠️ Token use adds up fast when your client explores multiple postings or fetches full job details, so consider a cheaper model.

Disclaimer: This is an unofficial, personal-use tool. It is not affiliated with, endorsed by, or sponsored by 104 Corporation, TSMC, or any other job board or company whose listings it searches. It calls each site's public web endpoints; please respect their terms of service and use it at a reasonable, low frequency. The project ships no scraped job data.

Install

With Homebrew:

brew install --cask amikai/tap/openings-mcp

With Go:

go install github.com/amikai/openings-mcp/cmd/openings-mcp@latest

Upgrade by rerunning the same command; pin a version with @vX.Y.Z.

Without Go: download the archive for your platform from Releases and put openings-mcp on your PATH.

With Docker (multi-arch: linux/amd64, linux/arm64):

docker pull ghcr.io/amikai/openings-mcp

Add the MCP server to your tool

With openings-mcp on your PATH:

Claude Code

claude mcp add openings-mcp -- openings-mcp

Codex

codex mcp add openings-mcp -- openings-mcp

Gemini CLI

gemini mcp add openings-mcp openings-mcp

With Docker instead, replace openings-mcp with docker run -i --rm ghcr.io/amikai/openings-mcp, e.g.:

claude mcp add openings-mcp -- docker run -i --rm ghcr.io/amikai/openings-mcp

Disclaimer

This is an unofficial tool. It is not affiliated with, endorsed by, or supported by 104 Corporation, TSMC, or any other job board or company whose listings it searches.

Some providers rely on undocumented APIs that may change or stop working at any time without notice. Job listing data belongs to the respective sites. The server fetches it on your behalf when you invoke a tool and stores or redistributes nothing.

Use this tool for personal job searching at a human pace. Every request goes out from your own machine, and a site may throttle or temporarily block your IP if too many arrive too quickly. The server does not rate-limit requests, so a client that pages through many results or fetches lots of job details in one go can trip these limits. The server never logs in, so your accounts on these sites are not at risk. You are responsible for complying with the terms of service of each site you query; do not use it for bulk scraping or data harvesting.

Credits

Inspired by these MCP job-search servers:

Company roster data cross-referenced from:

  • OpenPostings β€” company/ATS discovery data used to expand the curated roster

Jobindex search reference:

License

MIT