Integrations
An integration connects your existing AI client to Archestra. Model requests go through the LLM Proxy for authentication, logs, costs, and guardrails. Clients that support MCP can also use the MCP Gateway to call the tools assigned to an Archestra agent or gateway.
Choose the guide for your client:
- Coding clients: start with Connect Your Agents for generated setup. Claude Code covers subscription and API-key inference.
- Claude Desktop: use its setup helper, including conversation import and recovery.
- Manual model connections: follow n8n, Open WebUI, or the SDK and framework guides below.
- Hosted tools: connect a Microsoft Foundry agent to an MCP Gateway. Foundry continues to host the model.
An LLM virtual key authenticates model requests. A gateway token authenticates MCP calls. They are separate credentials; use each with its matching endpoint.
Explore
Claude Code
Connect Claude Code with subscription or API-key inference
Claude Desktop
Route Claude Desktop's inference and tools through Archestra
n8n
Connect an n8n AI workflow to Archestra
Open WebUI
Connect Open WebUI models and tools to Archestra
Vercel AI SDK
Route AI SDK model requests through Archestra
Pydantic AI
Route Pydantic AI model requests through Archestra
Mastra
Route a Mastra agent through the LLM Proxy
Microsoft Foundry
Connect a Foundry agent to Archestra MCP Gateway tools