Start a search

Starts a signal search and answers 202 with the search. Poll GET /searches/{search_id}, or pass ?wait=1 to 25 to wait that many seconds. Finished in time, the call answers 200. A raw search that succeeded within wait also inlines its rows in results.

Mode raw, the default of every type, returns the rows of the source and charges per row: a funding company is 1 credit, a technology company is 20, a person from engagement, listening, an event or a new role is 5. No AI judges a raw row. For the LinkedIn signals a row is a person as LinkedIn shows them next to a post (name, headline, what they did, the post), and you decide who fits: see Recipes. Mode screened has our AI check people against a buyer profile and charges 20 per exact lead and 5 per lead just outside. A search that delivers nothing is free.

Fields that belong to mode screened (buyer_profile, max_leads, just_outside, signal.about, signal.drop_intents) are 400 with code param_not_allowed_in_raw in mode raw; without mode, the message says to send "mode": "screened". An unknown technology is 400 with code unknown_technology and suggestions. In mode screened, an event name stops at requires_action so you choose the event. Mode raw never looks the event up: send its exact name with the year, its hashtags or its organizer page.

Authentication

Send Authorization: Bearer gtk_…. Create the key under API Access → API Keys. A missing, invalid or revoked key is 401, including while this API is closed. A valid key while the API is closed for your organization answers 404, the same body as an unknown path. Every response includes FounderMCP-Api-Version and FounderMCP-Request-Id.

Query parameters

• wait integer — Seconds to wait before answering, 0 to 25. Defaults to 0, which answers 202 at once.

Headers

• Idempotency-Key string · required — Required on this call. The same key and the same body replay the first response for 24 hours (Idempotent-Replayed: true). The same key with a different body is 409 idempotency_conflict.

Body parameters

• signal object · required — Required. type is linkedin_engagement (needs company.linkedin_url), social_listening (needs query), funding_raised (needs countries), new_in_role (needs titles), events (needs event.event_id or event.name; event_posts is the old name) or tech_adoption (needs technologies, raw only).
• mode string — Left out uses the type default: raw for every type. raw returns the rows of the source, with no AI judging them. Send screened to have our AI check people against buyer_profile (not for tech adoption). One of: raw, screened.
• buyer_profile object — Buyer profile, mode screened only. roles (departments: ceo, sales, marketing, finance, operations, hr, it, talent_acquisition, engineering, product, customer_success, data, procurement, legal), titles, levels (c_level, vp, director, manager), countries (ISO alpha-2, or EUROPE), allow_recruiters, allow_vendors. sectors and company_sizes only on new_in_role. Empty means anyone.
• exclude_companies object[] — Up to 20 LinkedIn company pages whose rows are never delivered. Tech adoption uses signal.exclude_domains instead.
• max_leads integer — Mode screened. Caps at 10 for engagement, listening and events, and 25 for funding and new-in-role. A higher number is clamped. Defaults to 25, then clamped.
• just_outside string — Mode screened. auto (default) and always fill up to 10 rows with near-misses. never returns exact leads only. One of: auto, never, always.
• limit integer — Mode raw. How many rows. Capped at the type maximum from GET /signals.
• cursor string | null — Mode raw. result.next_cursor of an earlier raw search with the same signal. One chain delivers at most 500 rows, kept 30 days.
• webhook_endpoint_id string | null — A webhook endpoint id (whe_…) that receives this search’s events.
• metadata object — Up to 10 string values. Keys 1–40 characters, values up to 200.
• max_credits integer | null — Stop before the next row would pass this. Below one row the call is 402 max_credits_exceeded and nothing runs.

Response

• id string · required — Search id, srch_….
• object string · required — Always search. One of: search.
• status string · required — Where the search is. One of: queued, running, requires_action, succeeded, failed, cancelled, expired.
• mode string · required — raw or screened. One of: raw, screened.
• signal_type string · required — The signal you asked for. One of: linkedin_engagement, social_listening, funding_raised, new_in_role, events, tech_adoption.
• progress object | null — stage and message while it runs.
• result object | null — Set when status is succeeded. Screened: exact, just_outside, total_estimate, max_leads_applied. Raw: rows, has_more, next_cursor.
• credits object · required — charged, reserved (worst case while open, 0 when finished) and max_credits.
• required_action object | null — Set on requires_action: type choose_event, candidates (up to 5) and expires_at (24 hours).
• error object | null — Set when status is failed: type, message, retryable.
• request object · required — The body as validated, with defaults filled in.

Credits

Charged per delivered row or lead, at the price in GET /signals. An exact screened lead is 20 credits. A lead just outside is 5. A funding company is 1. A technology company is 20. Nothing delivered, a failure, a cancel and an expiry are free. Errors and empty lookups are free. A call that spent credits puts that number in credits_used, in X-Credits-Used and in X-Credits-Charged. When the wallet can be read, the response also sends X-Credits-Remaining.

Errors

• 400 invalid_request — The body is not valid JSON, a field is unknown or has the wrong type, or a value is not allowed.
• 401 authentication_error — The Authorization header is missing, or the key is unknown or revoked.
• 403 permission_error — API access is not part of your plan, or your account is suspended.
• 404 not_found — The endpoint does not exist.
• 429 rate_limit_error — More than 60 requests per minute, or more than 3 at the same time, on one key. Wait for the Retry-After header.
• 429 daily_limit_reached — The request could take your organization over its daily credit limit. The limit resets at midnight (Europe/Brussels).
• 402 insufficient_credits — The credit balance is lower than this request could cost. Nothing was charged. Add credits, or send a smaller request. The error includes code, fix and a link to the credits docs.
• 502 upstream_error — The data source did not answer. Retry later.
• 503 service_unavailable — The API is temporarily unavailable. Retry later.
• 500 internal_error — Something went wrong on our side.
• 402 max_credits_exceeded — max_credits is lower than the credits of the first row, so nothing could be returned. Nothing was charged; raise max_credits or ask for fewer rows.
• 409 idempotency_conflict — The Idempotency-Key was already used with another request body.
• 409 idempotency_in_progress — The first request with this Idempotency-Key is still being processed. Wait for the Retry-After header.
• 422 unresolvable_input — A company, person or event in the request does not exist or cannot be found.

Example

cURL

curl -X POST "https://foundermcp.dev/api/v1/searches?wait=25" \
  -H "Authorization: Bearer $FOUNDERMCP_API_KEY" \
  -H "Idempotency-Key: idem-1" \
  -H "Content-Type: application/json" \
  -d '{
    "signal": {
      "type": "linkedin_engagement",
      "company": {
        "linkedin_url": "https://www.linkedin.com/company/acme-analytics"
      },
      "leads_from": "reactors"
    },
    "limit": 25,
    "max_credits": 125
  }'

JavaScript

const response = await fetch('https://foundermcp.dev/api/v1/searches?wait=25', {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.FOUNDERMCP_API_KEY}`,
    'Idempotency-Key': 'idem-1',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    "signal": {
      "type": "linkedin_engagement",
      "company": {
        "linkedin_url": "https://www.linkedin.com/company/acme-analytics"
      },
      "leads_from": "reactors"
    },
    "limit": 25,
    "max_credits": 125
  })
});
const body = await response.json();

Python

import json
import os
import urllib.request

payload = {
  "signal": {
    "type": "linkedin_engagement",
    "company": {
      "linkedin_url": "https://www.linkedin.com/company/acme-analytics"
    },
    "leads_from": "reactors"
  },
  "limit": 25,
  "max_credits": 125
}

request = urllib.request.Request(
    "https://foundermcp.dev/api/v1/searches?wait=25",
    headers={
    "Authorization": f"Bearer {os.environ['FOUNDERMCP_API_KEY']}",
    "Idempotency-Key": "idem-1",
    "Content-Type": "application/json"
    },
    data=json.dumps(payload).encode(),
    method="POST",
)
with urllib.request.urlopen(request) as response:
    body = json.load(response)

Example response

{
  "id": "srch_01J9Z6P8C8V0K3W6T2N8R5M1XA",
  "object": "search",
  "status": "succeeded",
  "mode": "raw",
  "signal_type": "linkedin_engagement",
  "funnel": {
    "posts": 10,
    "people": 23
  },
  "result": {
    "rows": 18,
    "has_more": false,
    "next_cursor": null,
    "total_estimate": 18,
    "truncated_by": null
  },
  "credits": {
    "charged": 90,
    "reserved": 0,
    "max_credits": 125
  },
  "sources": [
    "linkedin"
  ],
  "required_action": null,
  "error": null,
  "notes": [
    "Left out, and not charged: 5 people who work at Acme Analytics (by their headline).",
    "Raw rows are not checked against a buyer profile and no AI judged them: decide yourself who fits, from each headline and post. Send mode screened to have FounderMCP check people against a buyer_profile instead."
  ]
}