# Swiss Jobs API & MCP

Swiss job data for your AI and tools

Search Swiss jobs and employers in your AI or applications. Connect Claude, ChatGPT or Gemini CLI through MCP, or use the REST API. Start without an account.

## Choose your connection

[Compare Swiss job-data APIs and feeds](https://withpanta.com/data/swiss-jobs/compare/): a dated table and detailed comparisons with nine providers, including public search, MCP, publisher feeds and historical datasets. [Read the comparison as Markdown](https://withpanta.com/data/swiss-jobs/compare.md).

Claude, ChatGPT and Gemini CLI use the same remote MCP server for public job and employer data:

```text
https://withpanta.com/mcp
```

### Claude

1. In Claude, open Customize → Connectors and add a custom connector named Panta.
2. Paste the MCP server URL below. Choose No sign in for public job and employer searches.
3. Enable Panta in your conversation’s connectors menu, then try the example prompt.

Menus and availability depend on your Claude plan and organization. [Official setup guide](https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp).

### ChatGPT

1. On ChatGPT’s website, open Plugins, select the plus button, then Add custom MCP server.
2. Name it Panta, paste the MCP server URL below and choose No authentication for public data.
3. Create and install the plugin. In a conversation, type @ and choose Panta, then try the example prompt.

Your workspace permissions and security settings can affect access to custom MCP plugins. [Official setup guide](https://developers.openai.com/api/docs/guides/custom-mcp-server).

### Gemini CLI

1. Add Panta as a remote HTTP server with the terminal command below.
2. Run gemini mcp list to check that Panta is connected.
3. Ask Gemini CLI about Swiss jobs and employers, then inspect the sources in its results.

```sh
gemini mcp add --transport http panta https://withpanta.com/mcp
```

These instructions are for the terminal-based Gemini CLI. [Official setup guide](https://geminicli.com/docs/tools/mcp-server/).

Try: “Find software engineering jobs in Zürich. Show the employer, location evidence, original source link and when each listing was last checked.”

[Setup and available tools](https://withpanta.com/docs/mcp/). Personal workspace tools are optional and require separate OAuth consent.

## Use the REST API

```sh
curl --fail --get 'https://withpanta.com/api/v1/jobs' \
  --data-urlencode 'q=software engineer' \
  --data-urlencode 'canton=ZH' \
  --data-urlencode 'limit=3'
```

[Run the example request](https://withpanta.com/api/v1/jobs?q=software%20engineer&canton=ZH&limit=3) or [read the quickstart](https://withpanta.com/docs/quickstart/). Responses include pagination; follow next_offset. No account or key is required for public reads.

## What you get

### Jobs and employers

Search roles, locations and employers. Read full job details and the source postings behind them.

### Original sources

Keep the original listing and application links. Follow the employer’s source before acting on a result.

### Swiss location evidence

Distinguish a Swiss workplace, remote eligibility from Switzerland and roles spanning several countries.

### Observation dates

Inspect when a listing was observed and the source’s freshness. A checking date is different from a publication date.

## Inspect the evidence

Read source_url, location_class, observed_open_at and source_freshness. Unknown information stays unknown. [Understand the data](https://withpanta.com/docs/evidence/) and inspect [Panta’s original research](https://withpanta.com/insights/). Current coverage counts are available at https://withpanta.com/api/market-summary; counts describe the changing index, not national vacancy totals.

## Questions

### Do I need a Panta account?

Public job and employer searches work without a Panta account or API key, through both REST and MCP. A Panta account is only needed for optional API keys or personal workspace access.

### What does access cost?

Public market reads are currently available without a paid plan. Anonymous access allows 30 operations per minute per IP, per Cloudflare location. Optional API keys allow 60 operations per minute per account, shared with MCP connections. These are approximate abuse limits, not a guaranteed throughput.

### Can I use a different assistant or programming language?

Use a client that supports remote MCP over Streamable HTTP, or call the REST API from any language that can make HTTPS requests. Client features and organization policies can affect connector setup.

### How complete and current is the data?

Panta is an observed job index, not a census of the Swiss labour market. Sources refresh at different times and may become stale or unavailable. Check source evidence and freshness on each result; an active record does not guarantee that an employer still accepts applications.

### Can I download the entire market or a hiring history?

The public API supports bounded, paginated searches. It does not provide a bulk export or historical time-series API. An employer’s past listings are incomplete observed records, not a complete hiring history.

### Can I republish job descriptions?

Public access does not grant a license to republish third-party job descriptions. Preserve source links and attribution, and review Panta’s terms and the original source’s rights before reusing content.

## Documentation and terms

[API reference](https://withpanta.com/docs/reference/) · [Access and limits](https://withpanta.com/docs/authentication/) · [OpenAPI](https://withpanta.com/api/v1/openapi.json) · [Terms](https://withpanta.com/terms/)