Skip to content

The open-source agent harness - the runtime layer that turns an LLM into a working agent.

1.4 KB
Raw Blame
---
title: "Quickstart"
description: "Render a working agent chat UI against a TrueForge server"
---

## Installation

<CodeGroup>

```shellscript npm
npm install @truefoundry/trueforge-ui @truefoundry/trueforge-sdk
```

```text pnpm
pnpm add @truefoundry/trueforge-ui @truefoundry/trueforge-sdk
```

```text yarn
yarn add @truefoundry/trueforge-ui @truefoundry/trueforge-sdk
```

</CodeGroup>

`react` and `react-dom` are required peer dependencies (version 18 or 19).

## Render the UI

Point `TrueForgeUI` at your TrueForge server:

```tsx App.tsx
import { TrueForgeUI } from "@truefoundry/trueforge-ui";

export default function App() {
  return (
    <div style={{ height: "100dvh" }}>
      <TrueForgeUI
        server={{
          type: "trueforge",
          baseUrl: "http://localhost:8790",
          // token: "<bearer token>", // only if the server has OIDC login enabled
        }}
        layout="sidebar"
      />
    </div>
  );
}
```

`baseUrl` is your harness API root (defaults to `/` for same-origin). Add `token` — a Bearer token for the SDK — only when the server has OIDC login enabled. That is enough to get a full agent chat with streaming, history, and tool calls.

<Frame caption="What `TrueForgeUI` renders — a full agent chat with streaming, history, and tool calls.">
  ![The TrueForgeUI component rendering an agent chat with an agent-steps panel and a streamed response](/images/agent-composer-with-library.png)
</Frame>