LinkedIn MCP server

How an agent would call it

Two steps: search returns refs, then data_fetch opens one of them. The arguments below are the arguments that were sent.

{
  "tool": "search",
  "arguments": {
    "query": "Satya Nadella",
    "sources": [
      "linkedin"
    ],
    "limit": 2,
    "runId": "<from run_begin>"
  }
}

{
  "tool": "data_fetch",
  "arguments": {
    "sourceId": "linkedin",
    "ref": {
      "id": "satyanadella"
    },
    "runId": "<from run_begin>"
  }
}

This runs against an account — send Authorization: Bearer <token> with the call. Get one at app.bedrocktap.com/dashboard/tokens; the free tier needs no card, and it reaches linkedin like every other source.

What it costs. 5 credits per search call and 5 per fetch, whatever the row count. The search above returned 2 people for 5 credits; the fetch cost 5.

What came back

{
  "ref": {
    "id": "satyanadella",
    "url": "https://www.linkedin.com/in/satyanadella",
    "title": "Satya Nadella"
  },
  "text": "Chairman and CEO at Microsoft",
  "fetchedAt": "2026-09-12T15:03:25.595Z"
}

The search rows carried little more than location; the profile is what the fetch is for. Its data carried full_name, first_name, last_name, headline, location, public_identifier, urn, website, created_date and the standing flags — is_hiring, is_open_to_work, is_creator, is_influencer, is_premium and is_top_voice.

Observed 2026-09-12. One call against a live source — ask again tomorrow and the rows move.

What a row contains

One row is one person: their profile username as the id, the profile URL, and their name. meta carries the headline, the location and the current company where the row states them. The username is what a fetch needs, which is why it is the id. This is the least predictable source in the catalogue and it answers one person at a time — for hiring across a whole market, the job-postings source is the right call.

A question it answers. Is this founder still where they say they are, and is their company hiring?

Routing guidance

What an agent reads to decide whether linkedin is the right source, and what to reach for when it is not.

Answers

Professional profiles, hiring signals at a named company, and B2B audience research.

Query shape

A person’s name, a company name or a job keyword — people, jobs and posts are searched in parallel and the results merged.

Price

Priced in credits, per row, per query or per call depending on the source. Every call is quoted before it runs — `cost_estimate` returns the exact figure and spends nothing. See /pricing.

Legal basis

Public web — collected from pages published without a login.

Latency

Fast (~1-3s).

Prefer instead

`jobs` for job search — it aggregates several boards. Come here for profiles and company-specific detail.

Cannot answer

Private profiles, connection graphs, direct messaging, or anything requiring a logged-in session.

Related sources