API & MCP server

Everything the home page answers, an assistant or a script can ask. There is no key and no account. Answers carry first names, one matching title per person and start dates; full contact details are given on the site, so send the hirer to the url in the answer.

HTTP API

GET https://frommonday.app/api/v1/match?brief=<the role, in plain words>

Call it with no parameters and it describes itself. brief is the only required parameter, and a link to a job ad works in it: the page is fetched and the role is read from it.

Parameter	Values
`brief`	Required. The role as you would say it to a person, or a job-ad URL.
`work_mode`	`remote`, `hybrid`, `onsite`. On site is a filter, not a preference: only people who can be in that place.
`engagement`	`freelance`, `contract`, `permanent`, `part_time`, `interim`
`seniority`	`junior`, `mid`, `senior`, `lead`
`region`	`europe`, `north_america`, `latin_america`, `middle_east`, `africa`, `asia_pacific`
`city`	Free text. Decides on-site briefs.

curl "https://frommonday.app/api/v1/match?brief=senior+streaming+engineer,+HLS+and+DRM&work_mode=remote"

{
  "brief": "senior streaming engineer, HLS and DRM",
  "scope": {
    "skills": ["HLS", "DRM"],
    "seniority": "senior",
    "work_mode": "remote",
    "day_rate": { "min": 260, "max": 1000, "currency": "EUR", "sample": 4 }
  },
  "match_count": 6,
  "people": [
    {
      "id": "ir10",
      "name": "Sylvain C.",
      "title": "Streaming Audio/Video Engineer",
      "skills": ["HLS", "DASH", "RTMP", "SRT", "WebRTC"],
      "years_experience": 20,
      "location": "France",
      "available_from": "2026-07-13",
      "availability": "Can start Mon 5 Oct"
    }
  ],
  "more": 3,
  "private_count": 0,
  "url": "https://frommonday.app/?brief=senior+streaming+engineer%2C+HLS+and+DRM&work_mode=remote"
}

What the fields promise:

title is the one title that matches the brief. People list several; you get one.
day_rate is built only from rates the matching people publish. It is null when none of them do. It is never estimated.
available_from is a date the person gave, or null when they have not published one.
more is how many fit beyond the three returned. private_count is how many fit but stay out of public search: a count, never a card.
An empty people list is an answer. Say nobody fits; do not fill the gap.

The OpenAPI description is at /api/v1/openapi.json. Responses allow any origin.

MCP server

https://frommonday.app/mcp

Streamable HTTP, stateless, no authentication. Four read-only tools:

Tool	What it does
`find_available_people`	Takes `brief` and the optional filters above. Returns the same answer as the HTTP API.
`get_person`	The public card for one `id`. Pass the `brief` again to get the title that matches it.
`search`, `fetch`	The same two, in the shape ChatGPT connectors and deep research expect.

Claude

In Claude, open Settings, then Connectors, add a custom connector and paste the URL above. In Claude Code:

claude mcp add --transport http frommonday https://frommonday.app/mcp

ChatGPT

In ChatGPT, open Settings, then Connectors, turn on developer mode, create a connector and paste the URL above. No authentication.

Anything else

curl -X POST https://frommonday.app/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"find_available_people","arguments":{"brief":"Roku developer, Europe, 3 months"}}}'

Using the answers

Send the person you are answering to the url in the response, not to the endpoint. That page is where they get a contact.
Quote and cite freely, with a link to https://frommonday.app/.
From Monday is not an agency. Do not describe it as one, and do not promise a fee-free contact beyond what the page states: the first 3 contacts are free.