Live research API

Reliable, cost-efficient bulk enrichment — from live research.

Enrich leads & companies at scale. Validated, source-grounded JSON — via MCP or API.

Start for free Book a demo Read the API docs

Reliable at bulk scale
A fraction of the cost
Structured JSON — MCP or API
Built in Europe · GDPR-compliant

POST /research · ReAct trace live

TASK Find Series B SaaS in Berlin hiring Heads of Sales (schema: company, role, source_url).

THOUGHT Start by listing fresh Series B raises in Berlin.

ACTION NewsSearch("Berlin SaaS Series B 2026")

OBS 14 results — Tymyo, Pliant, Glean.io, Stark, …

THOUGHT Check careers pages in parallel.

ACTION DeepSearch × 5 (Tymyo, Pliant, Glean.io, Stark, Lemon.io)

OBS 3 open Head-of-Sales roles with sources.

FINAL [{"company":"Tymyo","role":"Head of Sales EMEA","source_url":"…"}, …]

200 · 8 tool calls · 4.2 s schema-valid

Token bill for GTM research: −80%
Domain in, validated JSON out: 1 call
Repeat companies: €0

Your list, enriched — column by column

Every column is a research question: Is the lead still there? Does the company fit your ICP? Who do they sell to? Any open sales role? The agent fills each cell with validated, source-grounded JSON — thousands of rows, ~80% less than running it on your own LLM tokens.

leads_q3.csv running column…

Company	AI Lead still there?	AI ICP match?	AI Sells to	AI Open sales role?
Tymyo tymyo.com	Yes	Yes Series B · 120 ppl · DACH	Mid-market SaaS	1 open Head of Sales EMEA
Pliant pliant.io	Left gone since 03/2026	Yes	Fintech & banks	None
Glean.io glean.io	Yes	No too small (<10 ppl)	Data teams	2 open
Stark stark.co	Yes	Yes	Design agencies	None

One row as validated JSON schema-valid

{
  "company": "Tymyo",
  "domain": "tymyo.com",
  "lead_still_at_company": true,
  "icp_match": { "match": true, "reason": "Series B SaaS, 120 employees, DACH" },
  "sells_to": "Mid-market SaaS teams",
  "open_sales_roles": [
    { "title": "Head of Sales EMEA", "url": "https://tymyo.com/careers/head-of-sales" }
  ],
  "source_urls": ["https://tymyo.com/about", "https://www.linkedin.com/company/tymyo"]
}

Built for research at scale

Every feature either cuts cost or makes data reliable — REST, MCP, structured outputs. No UI click required.

Thousands in parallel — without breaking

Enrich thousands of companies at once — reliably, with automatic retries on failure and webhooks on completion. Instead of fragile home-grown scrapers that fall over on volume and LinkedIn blocks.

Prebuilt GTM agents

Find leads, qualify companies, detect buying signals — ready-made agents for the most common GTM jobs. Or build your own with a custom JSON schema.

Into every GTM tool via MCP

Plug-and-play into Claude Code, Cursor, n8n, your CRM and any MCP client — research right where your GTM stack runs. No wrapper code.

Validated JSON — no hallucinations

JSON schema per column. The agent iterates until every required field is source-grounded, then formats exactly to your schema. Parseable, reliable, pipeline-ready.

Live research, ~1/10 the cost

Real live research across hard-to-access sources (LinkedIn, JS pages) instead of a stale database — at a fraction of the cost of a home-grown stack.

Built in Europe · GDPR-compliant

Built and hosted in Europe, fully GDPR-compliant. Optionally bring your own keys — clean for multi-workspace setups and high volume.

How it works

Three steps from prompt to validatable JSON.

Plug in — MCP or API

Add the MCP server to Claude Code or Cursor — or call POST /research. Two minutes, no wrapper code.

Send a company or domain

The agent researches 20+ sources live — including LinkedIn and JS-heavy pages. Results are cached, so repeats cost almost nothing.

Validated JSON returns

Optional JSON schema per request. The agent iterates until every required field is source-grounded — no hallucinations, straight into your pipeline.

POST /research

curl -X POST https://research-agent.net/research \
  -H "Content-Type: application/json" \
  -H "X-Auth-Token: $RA_KEY" \
  -d '{
    "prompt": "Find Series B SaaS companies in Berlin hiring Heads of Sales",
    "schema": "{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{\"company\":{\"type\":\"string\"},\"role\":{\"type\":\"string\"},\"source_url\":{\"type\":\"string\"}}}}"
  }'

Research Agent vs. the alternatives

Tavily, Exa, Serper and Firecrawl are pieces of the pipeline — you build the rest yourself. Clay is pricey and UI-heavy. We hand back the finished, structured company — including LinkedIn & JS pages that search-only APIs never see.

Provider	What it is	Cost / researched company	Output	LinkedIn & JS pages	Effort to a usable list	List from an ICP?	The catch
Research Agent	Research API / MCP tool	~€0.02–0.05	Validated JSON	Yes — incl. LinkedIn	Low	Yes — /build-list	—
Clay	GTM workbench (UI)	~€0.70–1.30	Structured (in-app)	Partial (integrations)	Medium (operate a UI)	Yes, but pricey	Expensive; complex credit math; you operate a UI, not an API
Claude / ChatGPT + web	Chatbot + web research	~€0.05–0.30*	Only if you enforce it	No (no LinkedIn)	High (you build it)	Ad hoc, not at scale	Looks free — but hallucinates, cannot read LinkedIn, does not scale to lists; done properly it costs more tokens than us
Tavily	Search / answer API	Primitive**	Raw text	No	High	No	Raw search results only — you build scraping, reasoning & qualification
Exa	Neural search API	Primitive**	Text / links	No	High	No	Search only — no scraping, no LinkedIn, no qualification
Serper	Google SERP API	Primitive**	Links only	No	Very high	No	Just Google links — you build scraping, enrichment & structure yourself
Firecrawl	Scraping API	Primitive**	Markdown, no reasoning	JS yes / LinkedIn no	High	No	Scrapes pages but no reasoning/qualification; LinkedIn blocked

* Your own unoptimized LLM tokens (raw HTML, no caching). ** Building blocks: €0.001–0.02 per call — but one company needs many calls + your own LLM + your own code. You end up rebuilding us.

Pricing

Prices that grow with you

One credit = one autonomous research run, no matter how many sources the agent hits. 400 free credits on signup.

Starter

~€0.025 per query

For solo builders, side projects, and first integrations.

€49 /month

or €124.95 per quarter (15% off)

or €441 per year (25% off)

2,000 credits / month

Autonomous ReAct research across 20+ sources
Structured JSON outputs
REST API + native MCP
Email support

Start with Starter

Growth

~€0.015 per query

For small teams piping data into real workflows.

€149 /month

or €379.95 per quarter (15% off)

or €1,341 per year (25% off)

10,000 credits / month

Everything in Starter
Bulk runs, processed in parallel
Custom agent templates + schemas
ICP scoring & buying signals
Priority support

Start with Growth

Scale

~€0.012 per query

For effectively unlimited volume + higher throughput.

€299 /month

or €762.45 per quarter (15% off)

or €2,691 per year (25% off)

25,000 credits / month

Everything in Growth
Bring your own model & search keys (BYOK)
Higher throughput limits
Webhook callbacks
Slack/Discord support

Start with Scale

All plans start with 400 free credits. No credit card. Cancel anytime.

Productive in 60 seconds

Start with 400 free credits. No credit card. Full MCP and REST API.

Start free Book a demo Read the API docs