Quickstart.

Four steps from a fresh account to a verified CSV. Counts and previews cost no contact credits, so you can size and inspect the market before anything is billed. Every step is shown twice, once for the CLI and once for the REST API, because they are the same engine.

1Get a key#

Create an account, then mint a key at app.argorant.com/profile → API keys. It is shown once and starts with ag_live_. The same key works on the REST API and the CLI.

cli

npm install -g argorant     # or use npx and skip the install
argorant login              # paste the key once

export ARGORANT_API_KEY=ag_live_…   # scripts and CI use the env var instead

Programmatic access is not on the Free plan.

The API, CLI, and MCP need either a paid plan or purchased pay-per-lead credits. Without one, every call returns 402 plan_required. Details on Authentication & limits.

2Count a market#

Always start here. A count tells you whether the segment is worth pursuing and costs no contact credits, so you can iterate on filters as long as you like.

cli

argorant count "fintech CFOs germany"
argorant count --industry fintech --title CFO --country Germany

curl

curl -G "https://app.argorant.com/api/mcp/people/count" \
  -H "Authorization: Bearer $ARGORANT_API_KEY" \
  --data-urlencode "industry=fintech" \
  --data-urlencode "title=CFO" \
  --data-urlencode "country=Germany"

# → { "ok": true, "count": 12480, "quota": { … } }

Free text in q is matched across name, title, company, and location; the structured filters narrow precisely. Both accept the same set: title, seniority, departments, industry, country, state, city, company_name, company_domain, has_email, has_phone, has_linkedin.

3Look before you buy#

A preview returns masked rows (initials, role, company, location) so you can judge whether the segment is actually who you meant. Still free.

cli

argorant search "fintech CFOs germany" -n 10

curl

curl -G "https://app.argorant.com/api/mcp/people/preview" \
  -H "Authorization: Bearer $ARGORANT_API_KEY" \
  --data-urlencode "industry=fintech" \
  --data-urlencode "title=CFO" \
  --data-urlencode "country=Germany" \
  --data-urlencode "limit=10"

4Export verified rows#

Exports are asynchronous: create the job, poll it, download the CSV. Deliverability is checked when the export runs, not when the database was built, so a scheduled job produces a sendable list every time.

cli

argorant export --industry fintech --title CFO --country Germany -n 500 -o cfos.csv

curl

# 1. create the job
curl -X POST "https://app.argorant.com/api/mcp/exports/create" \
  -H "Authorization: Bearer $ARGORANT_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: quickstart-001" \
  -d '{"filters":{"industry":"fintech","title":"CFO","country":"Germany"},"limit":500}'

# → { "job_id": 84213, "status": "queued", "status_api_path": "…" }

# 2. poll until it is done
curl "https://app.argorant.com/api/mcp/exports/84213" \
  -H "Authorization: Bearer $ARGORANT_API_KEY"

# → { "status": "completed", "total_rows": 500, "verified_rows": 463, "downloadable": true }

# 3. download the CSV
curl "https://app.argorant.com/api/mcp/exports/84213/download" \
  -H "Authorization: Bearer $ARGORANT_API_KEY" -o cfos.csv

Rather than polling, subscribe to the export.ready event and let Argorant call you, see Webhooks & idempotency. The Idempotency-Key above makes the create safe to retry.

Where to go from here#

You now have the whole loop: size, inspect, export. The rest of the docs are reference for doing it from somewhere other than a terminal.

REST API

Every endpoint, with runnable samples.

MCP

Hand the same loop to an AI agent.

Integrations

Zapier, Make, n8n, CRMs, senders.

No account yet?

Free tools run without one: look up a company, infer an email format, or verify a single address.

Create an account Company lookup Email format Email verifier

PreviousOverview NextAuthentication & limits