> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentity.to/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> What Agentity is and how it works.

Agentity gives AI agents a complete online identity. With a single API call your agent gets a real email address, a phone number, a single-use virtual payment card, and an encrypted secret vault — everything it needs to operate autonomously on the web.

Agents built on frameworks that support the [Model Context Protocol (MCP)](https://modelcontextprotocol.io) can connect directly to Agentity's MCP server without writing any REST integration code.

<Note>
  Agentity requires an active subscription. Without a subscription, identity provisioning calls will return an error.
</Note>

## Core concepts

<CardGroup cols={2}>
  <Card title="Identities" icon="id-card" href="/concepts/identities">
    An identity bundles email, phone, and virtual card credentials under a single agent ID.
  </Card>

  <Card title="Vault" icon="vault" href="/concepts/vault">
    An encrypted per-agent secret store for sensitive credentials like card numbers and tokens.
  </Card>

  <Card title="Providers" icon="plug" href="/concepts/providers">
    Third-party services (AgentMail, AgentPhone, Privacy.com) that issue the actual credentials.
  </Card>

  <Card title="Intents" icon="list-check" href="/concepts/intents">
    A structured audit log of every action your agent takes through Agentity.
  </Card>
</CardGroup>

## How it fits together

Each agent you build gets a unique `agentId` that you choose. When you provision an identity for that agent, Agentity calls the relevant providers on your behalf using the API keys you configured, then stores the resulting credentials. The vault lets your agent store sensitive data — like a one-time card number — so it can retrieve it later in the same session or across sessions.

All provider API keys and vault secrets are encrypted at rest. Agentity never stores plaintext credentials.

Use guardrails to control your agents' activities and prevent unexpected behavior.