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

# Agent-TARS

> Browser agent developed by the TARS team

Agent-TARS is a browser agent developed by the TARS team, distributed as an npm CLI package with powerful browser automation capabilities.

## Installation

```bash theme={null}
uv sync
npm install -g @agent-tars/cli@0.3.0
```

Activate `.venv` (or use `uv run bubench ...`) before running `bubench` commands.

<Note>
  Requires Node.js 18+.
</Note>

## Configuration

Configure Agent-TARS in the root `config.yaml` under `agents.Agent-TARS`:

```yaml theme={null}
agents:
  Agent-TARS:
    active_model: gpt           # active model profile
    models:
      gpt:                      # OpenAI model
        model_provider: openai
        model_id: gpt-5.4
        api_key: $OPENAI_API_KEY
        browser_control: hybrid
        timeout: 300
      claude:                   # Anthropic model
        model_provider: anthropic
        model_id: claude-sonnet-4-6
        api_key: $OPENAI_API_KEY
        browser_control: hybrid
        timeout: 300
      gemini:                   # Gemini via OpenAI-compatible endpoint
        model_provider: openai
        model_id: gemini-3-pro-image-preview
        api_key: $OPENAI_API_KEY
        base_url: $OPENAI_BASE_URL
        browser_control: hybrid
        timeout: 300
    browser:
      browser_id: local         # default local browser; see "Browser Modes" below to switch to cloud
```

Set `active_model` to the profile name you want to use by default, then switch at runtime with `--model <name>`.

### Config Parameters

| Parameter         | Description                   | Example                                    |
| ----------------- | ----------------------------- | ------------------------------------------ |
| `active_model`    | Default model profile         | `gpt`, `claude`                            |
| `model_provider`  | Model provider                | `openai`, `anthropic`, `volcengine`        |
| `model_id`        | Model ID                      | `gpt-5.4`, `claude-sonnet-4-6`             |
| `api_key`         | API key (use `$ENV_VAR` form) | `$OPENAI_API_KEY`                          |
| `base_url`        | API base URL (optional)       | `$OPENAI_BASE_URL`                         |
| `browser_control` | Browser control mode          | `dom`, `hybrid`, `visual-grounding`        |
| `browser_id`      | Browser backend               | `local` (default), `lexmount`              |
| `timeout`         | Task timeout (seconds)        | Default `300`, overridable via `--timeout` |

### Browser Control Modes

Agent-TARS exposes browser interaction as two independent tool sets — DOM tools and vision tools — that the model can select from. `browser_control` determines which tool sets are available:

| Mode               | How it works                                                              | When to use                                                                       |
| ------------------ | ------------------------------------------------------------------------- | --------------------------------------------------------------------------------- |
| `dom`              | Reads page structure and operates via DOM APIs (click, fill, etc.)        | Fast and stable for standard web pages with accessible DOM                        |
| `visual-grounding` | Takes a screenshot, uses a vision model to locate elements by coordinates | Handles canvas, dynamic content, and elements missing from the accessibility tree |
| `hybrid`           | Both tool sets exposed simultaneously; the model chooses per step         | Best fault tolerance — DOM is tried first, vision acts as fallback                |

`hybrid` is the recommended default for benchmarking.

## Browser Modes

**Local Browser**: Agent-TARS CLI launches a local browser. No extra parameters required.

```yaml theme={null}
browser:
  browser_id: local
```

<Warning>
  The Agent-TARS CLI launches Chromium itself and does **NOT** honour `local_proxy_*` (a one-time WARNING is logged if you set them). If you need to reach sites your bare connection can't (`bloomberg.com`, `archive.org`, etc.), switch to `browser_id: lexmount` — the international Lexmount region egresses outside mainland China. See [Local Chromium Browser](/en/browser/local) for the agent-by-agent support matrix.
</Warning>

**Lexmount Cloud Browser**: Connects to Lexmount via CDP. Apply for `LEXMOUNT_API_KEY` at [browser.lexmount.cn](https://browser.lexmount.cn) (mainland China) or [browser.lexmount.com](https://browser.lexmount.com) (international) — see [Lexmount Cloud Browser](/en/browser/lexmount) for the full flow — set it in `.env`, then reference it under `browser` in `config.yaml`:

```yaml theme={null}
browser:
  browser_id: lexmount
  lexmount_browser_mode: normal           # normal | light
  lexmount_api_key: $LEXMOUNT_API_KEY
  lexmount_project_id: $LEXMOUNT_PROJECT_ID
```

```bash theme={null}
# .env
LEXMOUNT_API_KEY=your-lexmount-api-key
```

To add a proxy:

```yaml theme={null}
browser:
  browser_id: lexmount
  lexmount_browser_mode: normal
  lexmount_proxy_server: http://gw.example.com:823
  lexmount_proxy_type: external
  lexmount_proxy_username: $LEXMOUNT_PROXY_USERNAME
  lexmount_proxy_password: $LEXMOUNT_PROXY_PASSWORD
```

## Usage

### Basic Run

```bash theme={null}
# Run first 3 LexBench-Browser tasks
bubench run \
  --agent Agent-TARS \
  --data LexBench-Browser \
  --mode first_n \
  --count 3

# Run Online-Mind2Web
bubench run \
  --agent Agent-TARS \
  --data Online-Mind2Web \
  --mode first_n \
  --count 3
```

### Run All Tasks

```bash theme={null}
bubench run \
  --agent Agent-TARS \
  --data LexBench-Browser \
  --mode all \
  --skip-completed
```

## Evaluation

```bash theme={null}
# --model-id matches the model_id used at run time
bubench eval --agent Agent-TARS --data LexBench-Browser --model-id gpt-5.4-mini

bubench eval --agent Agent-TARS --data Online-Mind2Web --model-id gpt-5.4-mini
```

## Supported Benchmarks

* ✅ LexBench-Browser
* ✅ Online-Mind2Web
* ✅ BrowseComp

## Related Links

* [Agent-TARS npm package](https://www.npmjs.com/package/@agent-tars/cli)