karriere.at MCP server: let your AI agent search Austrian job ads

#What your agent gets

The tool takes a required keywords input, for example python, buchhaltung or data scientist. The optional inputs are location and include_ads. location takes an Austrian city or state, such as Wien, Graz or Oberösterreich. include_ads defaults to false.

#Inputs and output

The result has keywords, location, total and jobs. total is the number of matches the page reports, and it can be null. Each job has title, company, place, employment_type, home_office, salary, posted, sponsored and url. The salary field carries the salary text the ad shows, and it is empty when the ad shows none.

The tool has no page input. Each call reads the first page of results, which holds about 15 job cards.

#How a request flows

1. You ask your agent: "Find Python jobs in Wien on karriere.at, with salary and link."
2. The agent calls find_recipe, then run with karriere.at/search-jobs, keywords set to python and location set to Wien.
3. wield-mcp passes the run through the local bridge to the Wield extension.
4. The extension sends the search request from your browser, so the recipe runs in your own browser and not on a server.
5. The job ads go back to your agent. Nothing passes through wield.run.

#What we saw in our check

On 7 November 2026, a search for python in Wien passed all 6 checks in one request, in 529 ms, and returned 132 matches. A data scientist search in Graz was also checked. A search with no matches gets a page with no job cards and an HTTP 404 status. The recipe reads that as zero results, not as an error.

#Sponsored ads

Ads with the label "Gesponsert" are paid placements. The recipe drops them by default, so the jobs list holds ordinary ads only. Set include_ads to true to keep them. Each kept ad has sponsored set to true, so your agent can tell them apart.

#What leaves your computer

The recipe makes one request, to www.karriere.at, for one search, as the site serves it to a visitor. The request leaves from your browser. Nothing passes through wield.run. Wield logs each request on your computer only, with method, host, path, status and timing, and never logs bodies or query values. Telemetry is off by default. Wield never reads, copies or sends your cookies or passwords.

#karriere.at's terms, in plain words

The site's robots.txt does not disallow anything for general agents. It has no rules for Claude agents or AI content signals. The recipe makes one request per call. It does not submit applications, set up job alerts or change a watchlist, and it makes no purchases. Our review of karriere.at's terms is not finished, and this is not legal advice. Use the recipe for one search that you start yourself, in your own browser, and not for bulk collection of ads. Read the site's terms yourself before you use this recipe in any other way.

#Limits and what is not working yet

• Read only. The recipe searches. It does not apply, set up job alerts or change a watchlist.
• First page only. The tool has no page input, so each call returns the first page of cards.
• Salary as stated. The salary is the text the ad shows. Many ads show none.
• Run limits apply. A run has at most 12 steps, 40 requests, 5 MB per response and 60 seconds.

#FAQ

#Can the agent apply for a job?

No. The recipe only searches. Open the ad's url to apply on karriere.at yourself.

#Does it need a karriere.at login?

No. The search works without an account.

#Does the search show every salary?

No. The salary field holds the salary text only when the ad states one.

#Are sponsored ads included?

Not by default. Set include_ads to true to keep them. They are marked with sponsored set to true.

#Which AI apps work with it?

Any MCP client. npx -y wield-mcp init sets up Claude Desktop, Claude Code, Cursor and Codex. Other MCP clients work with the standard stdio config.

#Where do the results go?

To your agent, on your own computer. The results never pass through wield.run.