External Tools#
Connect external AI tools to OptScale AI through a governed access layer. All supported integrations use the same OpenAI-compatible LLM proxy (llm_proxy/<organization_id>/v1) and an employee virtual key from AI Access, referred to as the API key throughout this guide.
Before you start, complete baseline setup in First Steps (organization, provider, etc). For Admin UI context, see Core Services — AI Access and Providers. To migrate existing application code from a direct vendor API or another gateway, see Switch to OptScale AI Gateway.
| Integration | Guide |
|---|---|
| OpenCode | Connect OpenCode |
| Cursor | Connect Cursor |
| Claude Code / Claude Desktop | Connect Claude |
Get connection values from Chat#
Use Chat to obtain the Base URL, API key, and Model name required for your integration. The dialog displays the values expected by the LLM proxy for your current selection, so there is no need to construct the URL or model identifier manually.
Steps
- Open Chat. In the top header, select the Provider and Model you plan to use in the external tool.
- Click Connect this model to external tools
in the left sidebar—see Interface Overview — Connect this model to external tools. - Open the tab corresponding to your integration. Use Base URL, API key, and Model name when set setitngs of the connection.
Model names#
A model name uniquely specifies a model available through OptScale AI. Examples:
anthropic/claude-sonnet-4-0
anthropic/claude-opus-4-0
openai/gpt-4.1
openai/gpt-4o
ollama/qwen3:32b
ollama/deepseek-r1:70b
Model names are used to configure external tools and clients. For connection instructions, see Cursor, OpenCode, and Claude Code and Claude Desktop.
To find an available model name, open OptScale AI Chat, select the required provider and model, and then click Connect this model to external tools. Open the tab corresponding to your integration and copy the value shown in the Model name field.
Shared Limitations#
All external tool integrations use the OptScale AI Gateway and therefore share the same requirements and limitations.
- The gateway URL must be accessible over HTTPS and use a trusted TLS certificate.
- Some features may require models that support specific capabilities, such as tool calling. Available functionality depends on the selected model.
- When policies are configured, requests made through external tools are subject to the same policy evaluation and enforcement as requests made through the Chat interface.
- Access to providers and models is controlled by the user's permissions and allowed provider configuration.
Related documentation#
After connecting external tools, you may want to:
- Review the Chat connection control in Interface Overview — Connect this model to external tools.
- Check setup requirements in First Steps.
- Manage virtual keys and user access in Core Services — AI Access.
- Review provider setup in Core Services — Providers.
- Validate policy behavior with End-to-end example — Policies + Guardrails.
- Understand the request path in Architecture Overview.