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

# Introduction

> Orca is a hosted platform for building, running, and publishing AI agents — with isolated workspaces, spend controls, and end-to-end observability built in.

## What is Orca?

Orca is a **hosted, multi-tenant platform for AI agents**. You define a reusable agent once — its
runtime, model, instructions, tools, skills, and memory — then run it, orchestrate it with other
agents, and publish it as a chat endpoint your own apps can call. Every run streams live, every
workspace is isolated per tenant, and usage, spend caps, and observability come standard.

Under the hood Orca separates a stateless control plane (the **Conductor**) from stateful session
management (**Runners**), so the platform scales horizontally to coordinate many concurrent agents
across multiple LLM providers. You never operate that split on the hosted product — it's what keeps
your agents fast and reliable.

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Create your account, build your first agent, and run it — in a few minutes.
  </Card>

  <Card title="Dashboard Guide" icon="table-columns" href="/dashboard/overview">
    A screen-by-screen tour of everything you can do in the Orca dashboard.
  </Card>

  <Card title="Core Concepts" icon="book" href="/concepts/profiles">
    Agents, sessions, runs, pools, workflows, skills, memory, and more.
  </Card>

  <Card title="SDKs" icon="box" href="/sdk/overview">
    Typed TypeScript, Python, and Go clients over the same API.
  </Card>
</CardGroup>

## Getting started

* **Sign up** at [www.orcapods.ai](https://www.orcapods.ai) and create (or join) an organization.
* **Use the dashboard** at [app.orcapods.ai](https://app.orcapods.ai) to build and run agents
  visually.
* **Call the API / SDKs** with an `ao_` tenant key (base URL `https://api.orcapods.ai`) for
  application code and automation.

<Note>
  Orca is also **self-hostable** — the source lives on GitHub and runs on Docker Compose or
  Kubernetes. See [Local development](/guides/local-development) and the deployment guides. The rest
  of this documentation leads with the hosted product; self-hosting notes are called out where they
  differ.
</Note>

***

## What you can build

<CardGroup cols={3}>
  <Card title="Reusable agents" icon="robot">
    Define an agent as a **profile**: a runtime, a model, a system prompt, and the tools, skills,
    and MCP servers it can use. Launch it anywhere, as many times as you like.
  </Card>

  <Card title="Multi-provider" icon="layer-group">
    Run on **Claude**, **OpenAI**, **Google**, or **Groq** — chosen per agent, not per deployment.
  </Card>

  <Card title="Live streaming" icon="bolt">
    Every run streams progress, tool calls, and token usage over Server-Sent Events, with replay,
    in the dashboard and via the API.
  </Card>

  <Card title="Multi-agent workflows" icon="sitemap">
    Compose agents into **pools** and **DAG workflows**; a deterministic engine schedules and runs
    nodes in parallel.
  </Card>

  <Card title="Skills & memory" icon="brain">
    Give agents packaged **Skills** and a per-agent **Memory Bank** that persists useful knowledge
    across sessions.
  </Card>

  <Card title="Tools & MCP" icon="plug">
    Built-in platform tools plus any external **Model Context Protocol** server — including SaaS
    apps connected through OAuth.
  </Card>

  <Card title="Publish as chat" icon="globe">
    **Publish** an agent behind a stable chat endpoint and mint scoped API keys for your app.
  </Card>

  <Card title="Isolated files & storage" icon="folder">
    Each session gets an isolated **virtual filesystem**; a shared artefact store holds agent
    outputs.
  </Card>

  <Card title="Credits, caps & RBAC" icon="shield-check">
    Prepaid **credits**, per-tenant **spend caps**, usage metering, and role-based access control
    keep teams safe by default.
  </Card>
</CardGroup>

***

## How it works

```mermaid theme={null}
flowchart TB
  Dashboard["Dashboard / SDKs / API"]
  Conductor["Conductor<br/>Stateless control plane<br/>POST /api/runs · session creation · SSE fan-out"]
  RunnerA["Runner A<br/>Stateful · sessions + tools"]
  RunnerB["Runner B<br/>Stateful · sessions + tools"]
  WorkerA["agent-worker<br/>MODE=claude"]
  WorkerB["agent-worker<br/>MODE=vercel"]
  Anthropic["Anthropic"]
  Providers["OpenAI / Google / Groq"]

  Dashboard -->|"REST"| Conductor
  Conductor -->|"SSE /api/runs/{id}/stream"| Dashboard
  Conductor --> RunnerA
  Conductor --> RunnerB
  RunnerA --> WorkerA
  RunnerB --> WorkerB
  WorkerA --> Anthropic
  WorkerB --> Providers
```

1. A dashboard action, SDK call, or `POST /api/runs` request names an agent profile and a prompt.
2. The **Conductor** places a session on a capable **Runner**.
3. The Runner builds a session-scoped toolkit and MCP endpoint, then hands the run to its
   **agent-worker sidecar**, which drives the chosen LLM.
4. Run events stream back through the Conductor to every subscriber in real time.

See [Architecture](/concepts/architecture) for the full design.

***

## Supported runtimes

Each agent picks a **runtime**, which selects how its model is driven:

| Runtime  | Backend             | Providers                                     |
| -------- | ------------------- | --------------------------------------------- |
| `pi`     | Pi coding agent SDK | Multi-provider (`provider:model`) *(default)* |
| `claude` | Claude Agent SDK    | Anthropic Claude                              |
| `codex`  | OpenAI Codex SDK    | OpenAI Codex                                  |
| `vercel` | Vercel AI SDK       | Anthropic, OpenAI, Google, Groq               |
| `poly`   | Capability router   | Multi-provider, tool-routed *(advanced)*      |

`general` is accepted as a legacy alias for the `vercel` runtime.

Models are named `provider:model`, e.g. `anthropic:claude-haiku-4-5`,
`anthropic:claude-sonnet-4-6`, `openai:gpt-5.2`, or `google:gemini-2.5-pro`. Availability and
pricing follow the platform's live rate feed.

***

## Where to next

<CardGroup cols={2}>
  <Card title="Build your first agent" icon="rocket" href="/quickstart">
    The fastest path from sign-up to a running agent.
  </Card>

  <Card title="Learn the dashboard" icon="table-columns" href="/dashboard/overview">
    Every screen, explained.
  </Card>

  <Card title="Agent profiles" icon="id-card" href="/concepts/profiles">
    How agents are defined and configured.
  </Card>

  <Card title="Publish an agent" icon="globe" href="/concepts/publishing">
    Expose an agent as a chat endpoint for your app.
  </Card>
</CardGroup>