

Verify humans. Reuse identity.
Valyd gives developers two straightforward ways to add trusted human verification: check that someone is a live, unique human with a quick redirect to Valyd’s verification page, or let users connect their Valyd identity and reuse verified information — running any extra checks your app needs.
Machine-readable — hand this to your AI agent
docs.valyd.work/llms.txt
- llms-full.txt — every page in one file
- Agent integration guide — the decision flow
- API reference · OpenID config
Two products
Pick the outcome you need — each card takes you straight to working code.
Determine whether someone is live and unique with your API key — start a session, redirect to Valyd’s verification page, read the verdict. No user accounts.
Unique Human APIUsers connect their Valyd identity, share verified information they already hold, and complete any additional checks your app requires — KYC, licenses, face match, and more.
Reusable VerificationCommon tasks
Is this a live human in front of the camera? Valyd captures and scores it; the verdict comes back to you.
Check livenessStandard OpenID Connect — a two-line button connects a user’s verified identity (and can double as sign-in).
Connect with ValydConfigure the checks your app needs, start a session, read the decision.
Run a verification workflowInvite teammates with roles, share apps, and bill one account.
Organizations & teamsThe API Playground runs the real flow against demo users.
Try it liveThe Valyd MCP server turns verification into agent tools.
Build with AI agentsGet running in 3 steps
Create an app, copy your credentials, call the API — start from the result you need.
- 1
Create an app & copy credentials
An app in the developer portal gives you an API key for the Unique Human API and OIDC client credentials for Connect with Valyd.
- 2
Call the API or connect the user
Call the Unique Human API with just your API key, or use Connect with Valyd and run a configured verification workflow for the user.
- 3
Read the result or decision
Unique Human API results return directly to your call. Workflow decisions arrive via statuses and webhooks, with passed proofs saved to the user’s Valyd ID.
Going deeper
Security & data
How Valyd protects the data an integration touches, and where to find the formal assurances your security review will ask for.
For AI agents & MCP
Everything here is machine-readable — point your agent at the guide and the specs.
Agent integration guide — the decision flow your agent follows to pick a product and call the API. The full llms.txt index is in the panel at the top.
Open →Developer resources
Everything machine-readable, ready to download or point your tooling at.
Need a hand?
Email support@valyd.id or manage your apps and credentials in the developer portal.