# Introduction

**Maetra** keeps AI agents inside the lines at runtime. Anchor long-running work to an active **Task Guard** contract, put a **checkpoint** in front of sensitive actions, and put a **scanner** in front of prompts, tool calls, and outputs — over a plain REST API or an MCP server.

New to Maetra? Start with the [Quickstart](https://maetra.io/docs/getting-started/quickstart). Just need a reference? Jump to the [API reference](https://maetra.io/docs/reference/api-reference).

### The three runtime controls

#### Task Guard

**AI agent task alignment.** Turn the user's objective, constraints, scope, and success criteria into an active task contract. Before a material action, Task Guard checks whether the work is aligned, supporting, needs an explanation, requires user confirmation, or should be refocused. Afterward, it can compare expected and reported effects.

#### Govern

**Human-in-the-loop approvals.** Before an agent does something consequential - move money, delete data, email a customer - it calls Maetra. Your exact policies or AI agent decision intelligence decide: auto-approve, auto-block, or route to a human. The agent gets a decision plus a **signed token** it can present as proof the action was authorised.

#### Secure

**Runtime content scanning.** Screen prompts, tool calls, and model outputs against your workspace rules — prompt injection, data exfiltration, disallowed tools — and get back a verdict (`safe`, `flagged`, `blocked`) and a recommended action. Anything flagged or blocked becomes an incident.


### Two ways to integrate

<table data-view="cards"><thead><tr><th>Interface</th><th>Best for</th><th></th></tr></thead><tbody><tr><td><strong>REST API</strong></td><td>Backends and services calling <code>/v1</code> over HTTPS.</td><td><a href="https://maetra.io/docs/getting-started/quickstart">Quickstart →</a></td></tr><tr><td><strong>MCP server</strong></td><td>AI agents and assistants that speak the Model Context Protocol.</td><td><a href="https://maetra.io/docs/mcp-server/overview-and-connection">MCP overview →</a></td></tr></tbody></table>

Both are backed by the same evaluation engine and authenticated with the same workspace API key.

> **Good to know**
> **You don't need to register your agents first.** Every endpoint accepts a free-text `agent_name`, so you can integrate in minutes and register agents later for richer reporting. Govern still evaluates API-only and MCP agents against organization-wide policies; when `agent_name` or `agent_id` matches a registered agent, agent-scoped policies can apply. See [Agents](https://maetra.io/docs/agents).


### Base URL

```
https://api.maetra.io
```

All REST endpoints are versioned under `/v1`. The MCP server has its own base URL — see the [MCP overview](https://maetra.io/docs/mcp-server/overview-and-connection).

### Next steps

1. [Create an API key](https://maetra.io/docs/getting-started/authentication) in the dashboard.
2. [Run your first checkpoint and scan](https://maetra.io/docs/getting-started/quickstart), or [start a Task Guard task](https://maetra.io/docs/task-guard-api/tasks-and-context).
3. Browse the [full API reference](https://maetra.io/docs/reference/api-reference).