API infrastructure for AI agents.

Email verification
built for agents.

Our first product: one endpoint, five verification layers, and a response schema that stays stable. OpenAPI spec at /openapi.json — AI coding tools read it and safely write your integration. EU-hosted. No dashboards. No onboarding. Just the API.

More APIs coming — screenshots, PDFs, HTML→Markdown, and more.

MCP server available — connect Claude, Cursor, and other clients at mcp.truval.dev/mcp. Read MCP guide →

$ get api key read docs → mcp openapi.json openapi.json
<300msP99 latency
50k+blocked domains
EUhosted · GDPR
5verify layers

Get started in one paste

// starter promptpaste into Claude, ChatGPT, Cursor, …
Integrate Truval email verification.

— Base URL: https://api.truval.dev
— OpenAPI spec (read first): https://api.truval.dev/openapi.json
— Endpoint: POST /v1/email/verify
— Headers: Authorization: Bearer <TRUVAL_API_KEY>, Content-Type: application/json
— Body: {"email":"<address>"}

The user supplies TRUVAL_API_KEY from https://dash.truval.dev — never invent a key.

Use the response fields: valid, status, confidence, disposable, smtp_blocked, role, suggestion, latency_ms.

Agents & IDEs

request / response live processing
request.sh
curl https://api.truval.dev/v1/email/verify \
  -H "Authorization: Bearer sk_live_..." \
  -H "Content-Type: application/json" \
  -d '{"email":"user@example.com"}'
response.json · 187ms
{
  "email":      "user@example.com",
  "valid":      true,
  "status":     "deliverable",
  "confidence": 0.97,
  "disposable": false,
  "role":       false,
  "suggestion": null,
  "latency_ms": 187
}
// why truval
01
agent-native
OpenAPI 3.1 spec at /openapi.json. Your AI logic agent reads it, understands the parameters, and writes your integration flawlessly. No docs page to manually navigate.
02
five layers, early exits
Syntax → disposable → MX → SMTP → signals. Bad addresses fail in under 5ms. Real addresses get a confidence score, not just true/false.
03
honest about limits
Gmail blocks SMTP probing from every third party. We tell you that with smtp_blocked: true and a calibrated 0.75 confidence — not a fake "valid".
04
EU-hosted · GDPR
Data stored seamlessly in secure European data centers. Delivered globally with high availability and built-in redundancy layers.
05
typo detection
gnail.com → suggestion: "gmail.com". Agents can catch and resurface this logic to operators before a bounce happens. Edit distance checked.
06
usage-based pricing
Free tier, then pay per call. No seats, no plans. Just a transparent transaction layer built for developers scaling up infrastructure.
// unified response schema · POST /v1/email/verify
validbooleantrue if the address passed all requisite validation phases.
statusstringdeliverable · undeliverable · unknown · catch_all · invalid
confidencefloat0–1 scale. 0.97 = SMTP confirmed. 0.75 = domain real, provider blocks probing.
disposablebooleanMatched immediately against 50k+ updated throwaway domain nodes.
smtp_blockedbooleanTrue for providers like gmail, outlook, yahoo which block SMTP probing.
rolebooleanadmin@, info@, noreply@ — structurally deliverable but not a real person.
free_providerbooleangmail.com, yahoo.com, hotmail.com flag identifiers.
suggestionstring|nullAlternative syntax mapping, e.g. "gmail.com" when input was "gnail.com".
mx_foundbooleanMX lookup propagation flag exists for the target domain zone.
failed_checkstring|nullFailure state layer: syntax · disposable · no_mx · smtp.
latency_msnumberEnd-to-end routing processing time benchmark.

Integrate in 60 seconds

Always free tier — 500 verifications per month, no card footprint. Builder pipelines scale natively starting at €9/month when traffic increases.

$ get free api key read docs →
// pricing infrastructure
zero500 caps/mo · €0
builder5k caps/mo · €9
scale100k caps/mo · €49
overage€0.002 / node