> ## Documentation Index
> Fetch the complete documentation index at: https://opensandbox-oc-s-762ac928075c46d2828bcb22.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Overview

> Run AI agents on OpenComputer — pick a core, connect channels, install packages

A managed agent on OpenComputer is a **core** plus the **channels** and **packages** you bolt onto it. Pick a runtime, connect the messaging platforms you want, install the capabilities you need.

<Info>
  **There are two ways to run agents on OpenComputer.** This page covers **hosted agents** — platform-managed agent definitions you reach through persistent instances or ephemeral sessions, and connect to messaging channels or a public API. If you'd rather create a sandbox yourself and drive a Claude session inside it directly with the SDK, see the [Agent SDK](/agents/overview).
</Info>

```
ManagedAgent = Core + Channels + Packages + Secrets
```

## Cores

The AI runtime that powers your agent. OpenComputer ships two blessed cores:

| Core                                 | What you get                                                                                     |
| ------------------------------------ | ------------------------------------------------------------------------------------------------ |
| [`hermes`](/agents/cores/hermes)     | Self-improving agent with skill creation, memory, multi-channel gateway, browser, code execution |
| [`openclaw`](/agents/cores/openclaw) | Multi-channel AI gateway with plugin SDK, native MCP support, hot-reload config                  |

Picking a core is the one decision that matters most — it determines the agent's capabilities, config model, and extension surface.

## Packages

Stateful extensions beyond the core — installed via `oc agent install`. Packages can provision infrastructure (managed databases, vector stores) and wire it into the agent transparently.

| Package                             | What it adds                                                                                                              |
| ----------------------------------- | ------------------------------------------------------------------------------------------------------------------------- |
| [`gbrain`](/agents/packages/gbrain) | Persistent memory — 34 MCP tools for pages, vector search, graph, timeline, files. Backed by a managed Postgres database. |

## Channels

Messaging platforms connected to your agent.

| Channel                                 | How it works                                                                          |
| --------------------------------------- | ------------------------------------------------------------------------------------- |
| [`telegram`](/agents/channels/telegram) | Webhook-based. Messages flow through the OpenComputer gateway to your agent and back. |

## Three commands to a running agent

```bash theme={null}
oc agent create my-agent --core hermes       # boot an agent
oc agent connect my-agent telegram           # connect Telegram
oc agent install my-agent gbrain             # add persistent memory
```

## Two ways to interact

Everything you can do with the CLI you can do with the REST API, and vice versa.

<CardGroup cols={2}>
  <Card title="CLI" icon="terminal" href="/cli/agents">
    `oc agent` commands — create, connect, install, inspect
  </Card>

  <Card title="REST API" icon="code" href="/agents/rest/agents">
    HTTP endpoints for agents, instances, sessions
  </Card>
</CardGroup>

## Agents, instances, sessions

An **agent** is configuration. Its actual compute runs in one of two modes:

| Concept                                | What it is                                                      | Use case                               |
| -------------------------------------- | --------------------------------------------------------------- | -------------------------------------- |
| [**Agent**](/agents/rest/agents)       | Configuration — core, snapshot, entrypoint, secrets             | Define once, reuse                     |
| [**Instance**](/agents/rest/instances) | Persistent sandbox, hibernates/wakes, multiplexes conversations | AI coworkers on Slack, Teams, Telegram |
| [**Session**](/agents/rest/sessions)   | Ephemeral sandbox, runs to completion                           | Batch jobs, webhook handlers           |

Managed agents (those with a core) get an instance automatically on creation. Raw agents — which have no core and use your own `snapshot` and `entrypoint` — create instances or sessions explicitly via the REST API. Managed and raw agents coexist on the same platform.

## Authentication

All requests require an OpenComputer API key via `X-API-Key` header or `--api-key` CLI flag.

```
X-API-Key: osb_your_api_key
```

Base URL for REST: `https://api.opencomputer.dev`