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

# Use Cline with RunBridge AI

> Set up Cline with a RunBridge AI OpenAI-compatible endpoint, model, and capabilities.

[Cline](https://cline.bot) connects to RunBridge AI through its **OpenAI Compatible** provider in VS Code.

## Prerequisites

* VS Code and the official [Cline extension](https://marketplace.visualstudio.com/items?itemName=saoudrizwan.claude-dev).
* An active API key from the [RunBridge AI console](https://runbridge.ai/console/token).
* A [text model](/overview/models) available to your key that supports Chat Completions and streaming. Choose a model with tool calling for agent tasks, and image support only if you need image input.

## Install and configure Cline

Install Cline from the linked VS Code Marketplace listing, open its sidebar, and select the settings gear. Configure these fields:

| Setting | Value |
| - | - |
| API Provider | **OpenAI Compatible** |
| Base URL | `https://api.runbridge.ai/v1` |
| API Key | Your RunBridge AI API key |
| Model ID | The exact RunBridge AI model ID you want to use |
| Use Azure Identity Authentication | Off |

Keep `/v1` in the Base URL. The OpenAI-compatible adapter adds `/chat/completions` and authenticates with a Bearer token. RunBridge AI is the custom endpoint you enter, rather than a built-in Cline provider selection.

In **Model Configuration**, set **Context Window** and **Max Output Tokens** to the selected model's supported limits. Enable **Image Support** or **Computer Use** only when the model and its API route support them. Enter current input/output pricing if you want Cline's local cost estimate; it does not replace RunBridge AI billing.

If you use separate configurations for Plan and Act, choose the intended RunBridge AI provider and model in both modes.

## Verify the connection

Save the settings and start a new task in a test folder. Ask: **Reply with CONNECTED only. Do not edit any files or run commands.**

This makes a real API request and may incur usage. Check that a response arrives without authentication or endpoint errors. Then request a small coding change and review Cline's proposed tool actions before approving them. This checks agent behavior beyond basic text connectivity.

## Troubleshooting

| Symptom | Check |
| - | - |
| HTTP 401 | Re-enter your RunBridge AI key in Cline's API Key field and check that Azure authentication is disabled. |
| HTTP 404 | Use `https://api.runbridge.ai/v1`, without a trailing `/chat/completions`. |
| Model not found | Enter an exact model ID available to your key; display names and provider-prefixed aliases may differ from API IDs. |
| Context length or output limit errors | Reduce the configured limits to match the selected model and start a fresh task if its history is too large. |
| Image or tool failures | Check the model's capabilities and the matching Model Configuration fields; do not enable capabilities based only on the model name. |
| Plan works but Act fails | Check both modes' provider settings if separate Plan/Act models are enabled. |

See Cline's official [OpenAI Compatible guide](https://docs.cline.bot/provider-config/openai-compatible) for the current settings and advanced options.

<script type="application/ld+json">
  {`
    {
    "@context": "https://schema.org",
    "@type": "TechArticle",
    "headline": "Use Cline with RunBridge AI",
    "description": "Set up Cline with a RunBridge AI OpenAI-compatible endpoint, model, and capabilities.",
    "url": "https://runbridge.mintlify.site/integrations/cline",
    "publisher": {
      "@type": "Organization",
      "name": "RunBridge AI"
    }
    }
    `}
</script>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.