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

# In-Process Runtime

> Run the full agent loop in-process with no external infrastructure or Temporal required

The in-process runtime is the **default**. When you call [`NewAgent`](/getting-started/quickstart) without Temporal options, the agent loop runs directly inside your Go process. No Temporal server, no workers, no external orchestration.

## Enable

Omit all Temporal configuration:

```go theme={null}
a, err := agent.NewAgent(
    agent.WithLLMClient(llmClient),
    agent.WithSystemPrompt("You are a helpful assistant."),
)
if err != nil {
    return err
}
defer a.Close()

result, err := a.Run(ctx, "Hello", nil)
```

See [Quickstart](/getting-started/quickstart) for a complete example.

## How it works

When you call `Run`, `RunAsync`, or `Stream`, the SDK executes the full agent loop in your process:

1. Load conversation history and recalled memory (when configured)
2. Call the LLM with the merged tool list
3. Execute tool calls and append results
4. Repeat until the model finishes or the iteration limit is reached
5. Return the result or stream events back to your caller

Everything happens in-process — no network hops, no external orchestration.

## Capabilities

All SDK capabilities are available on the in-process runtime:

* LLM providers (OpenAI, Anthropic, Gemini, custom)
* Tools, MCP, A2A, sub-agents
* Conversation (in-memory), memory, retrieval (RAG)
* Streaming, AG-UI events, hooks
* Human-in-the-loop approvals
* OpenTelemetry observability

The only capabilities exclusive to Temporal are **durable execution** (crash recovery) and **horizontal worker scaling**.

## Limitations

| Limitation                      | Detail                                                                  |
| ------------------------------- | ----------------------------------------------------------------------- |
| **No crash recovery**           | If the process exits mid-run, the run is lost — no replay, no resume    |
| **Single process**              | No horizontal worker scaling                                            |
| **In-memory conversation only** | Redis works technically, but split-process deployments require Temporal |
| **No split client/worker**      | `DisableLocalWorker` and `NewAgentWorker` require the Temporal runtime  |

<Note>
  These limits matter for production workloads where runs must survive deploys or scale independently. For those cases, see [Temporal runtime](/runtimes/temporal).
</Note>

## When to use

| Scenario                                       | In-process fit |
| ---------------------------------------------- | -------------- |
| Local development and unit testing             | Ideal          |
| CI pipelines and integration tests             | Ideal          |
| Simple API or script with short runs           | Good           |
| Production chat with crash recovery required   | Use Temporal   |
| Multi-process or horizontally scaled execution | Use Temporal   |

## Switch to Temporal

Add `WithTemporalConfig` — no other code changes:

```go theme={null}
a, err := agent.NewAgent(
    agent.WithTemporalConfig(&agent.TemporalConfig{
        Host: "localhost", Port: 7233,
        Namespace: "default", TaskQueue: "my-app",
    }),
    agent.WithLLMClient(llmClient),
    agent.WithSystemPrompt("You are a helpful assistant."),
)
```

A running Temporal server is required. See [Temporal](/runtimes/temporal) and [Prerequisites](/getting-started/prerequisites).

## Examples

<CardGroup cols={2}>
  <Card title="Simple Agent" icon="play" href="/examples/simple-agent">
    Minimal in-process agent with Run()
  </Card>

  <Card title="Stream" icon="wave-square" href="/examples/stream">
    Streaming tokens and tool events in-process
  </Card>

  <Card title="Runtimes Overview" icon="scale-balanced" href="/runtimes/overview">
    Compare in-process and Temporal
  </Card>

  <Card title="Quickstart" icon="bolt" href="/getting-started/quickstart">
    Build your first agent
  </Card>
</CardGroup>