VerifiedProxy Is Now Live

Today, we're excited to announce that VerifiedProxy is officially live and open to everyone. You can sign up now, create your account, register your first AI agent, and start browsing the web with a verified identity — all in under five minutes.

We built VerifiedProxy to solve a fundamental problem: as AI agents increasingly act on behalf of organisations — browsing websites, collecting data, filling forms, interacting with services — there has been no standard way for the websites they visit to know who they are. Until now.

What VerifiedProxy Does

VerifiedProxy is the identity layer for the agentic web. It gives your AI agents a cryptographically verified identity that any website can confirm in real time, with a single API call — no credentials required on the website's side.

When your agent browses the web through VerifiedProxy, every request it makes carries verifiable proof of who sent it, which organisation it belongs to, and what it's authorised to do. Websites can check this instantly and decide how to respond — whether that means bypassing CAPTCHAs, granting higher rate limits, or providing preferential access to verified agents.

How It Works

The setup is deliberately simple. There are no complex integrations, no SDKs to embed, and no changes to your agent's code. Here's how you go from sign-up to verified agent in four steps:

1. Create your account Sign up at app.verifiedproxy.com and receive your organisation API key. This key links everything back to your verified organisation.
2. Register your agent Run npx verifiedproxy install in your terminal. The interactive wizard walks you through naming your agent and declaring its scopes — what it's authorised to do, such as browse, form-fill, or data-collect. Your agent receives a unique ID and a secret credential.
3. Start the proxy Run npx verifiedproxy proxy to start the lightweight local proxy on your machine. It sits between your agent and the internet, automatically attaching verified identity headers to every outbound request.
4. Browse with identity That's it. Your agent now browses the web with a verified, organisation-backed identity. Every website it visits can confirm exactly who it is and who sent it.

The entire process requires Node.js 18+ and takes less than five minutes. No global installation is needed — everything runs via npx.

Built for Claude Code — and Everything Else

If you're using Claude Code, VerifiedProxy works out of the box. The installer automatically configures your .claude/settings.json so Claude Code routes all traffic through the proxy. There is nothing else to set up.

But VerifiedProxy isn't limited to Claude Code. Any AI agent or tool that supports standard HTTP_PROXY and HTTPS_PROXY environment variables can use it. Custom agents built with the Claude Agent SDK, browser automation tools, web scrapers — if it makes HTTP requests, it can carry a verified identity.

What Websites See

For website owners, verification is free, public, and requires no sign-up. Any website can verify an incoming agent by calling a single endpoint:

// Verify an incoming agent — free, no API key required
GET api.verifiedproxy.com/api/agents/verify

{
  "agent_id": "ag_abc123",
  "session": "session_token"
}

———————————————————————

200  Verified

{
  "agent_name":   "Research Agent",
  "org_name":     "Acme Corp",
  "scopes":       ["browse", "data-collect"],
  "valid":        true
}

The response tells the website exactly which agent is visiting, which organisation it belongs to, and what it's authorised to do. From there, the website decides how to handle the request — grant access, apply custom rate limits, skip bot detection, or log the visit for analytics.

What You Can Do With It

VerifiedProxy opens up possibilities for both sides of the agentic web:

If you deploy AI agents

• Get past bot detection Websites that recognise verified agents can let them through without CAPTCHAs, JavaScript challenges, or IP-based blocking.
• Access higher rate limits Verified agents can receive preferential treatment — more requests, faster responses, and access to content that anonymous bots would never see.
• Build trust with the sites you interact with Your agents carry your organisation's identity. That means websites know you're a legitimate business, not an anonymous scraper.
• Manage multiple agents from one account Register as many agents as you need, each with different names, scopes, and credentials — all under a single organisation.

If you run a website

• Know who's visiting Identify AI agent traffic in real time, tied to real organisations — not just user-agent strings that anyone can fake.
• Implement scope-based access Grant different levels of access based on what an agent is authorised to do. A browse agent gets read access; a data-collect agent might get API access.
• Track and audit agent activity Log which organisations' agents are visiting your site, how often, and what they're doing — all verified and attributable.

Designed to Never Block Your Agents

We built VerifiedProxy with a fail-open design. If the VerifiedProxy API is ever unreachable or a session can't be created, the proxy forwards your agent's requests normally — without identity headers, but without interruption. Your agent's work is never blocked by our infrastructure. The proxy binds to localhost only, keeping your traffic secure and never exposed to the network.

Get Started Today

VerifiedProxy is live, and you can start right now. Creating an account is straightforward, registering your first agent takes minutes, and the documentation covers everything you need to get up and running.

Here's where to begin:

• Create your account Sign up and get your organisation API key.
• Read the quickstart guide Go from zero to verified agent in five minutes.
• Explore the full documentation Dive into agent registration, the proxy, website integration, and the API reference.