A lightweight Node.js proxy that enables seamless integration between Azure OpenAI and OpenClaw by resolving API versioning and URL path conflicts.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install azure-proxy
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install azure-proxy using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Azure OpenAI Proxy?
The Azure OpenAI Proxy is a specialized utility designed to bridge the gap between Azure's unique API requirements and the standardized OpenAI-compatible requests used by OpenClaw. Because Azure requires specific deployment paths and api-version query parameters, direct connections often fail with 404 errors. This skill acts as a local intermediary, rewriting requests on the fly so your Openclaw Skills can communicate with Azure resources without modification.
By deploying this proxy, developers can leverage enterprise Azure credits, such as those from Visual Studio subscriptions, to power their autonomous agents. It handles the translation of headers and URL structures, ensuring that OpenClaw sees a standard OpenAI endpoint while Azure receives the specific metadata it requires for authentication and routing.
Azure OpenAI Proxy Use Cases
- Using Azure OpenAI as the primary model provider for Openclaw Skills to utilize enterprise credits.
- Resolving 404 Resource Not Found errors caused by OpenClaw's default URL construction logic.
- Routing high-volume subagent tasks through Azure deployments to manage costs and rate limits.
- Integrating Azure-hosted GPT-4o models into automated development workflows.
How Azure OpenAI Proxy Works
- The proxy server starts a local Node.js instance (default port 18790) that listens for OpenAI-formatted API calls.
- When an agent sends a request, the proxy intercepts the standard /chat/completions path.
- The proxy reconstructs the URL to include the required Azure deployment name and api-version parameter.
- It strips standard Bearer tokens if necessary and injects the Azure-specific api-key header.
- The modified request is forwarded to the Azure OpenAI endpoint, and the response is passed back to the agent seamlessly.
Azure OpenAI Proxy Setup
1. Start the Proxy
# Configure your Azure environment
export AZURE_OPENAI_ENDPOINT="your-resource.openai.azure.com"
export AZURE_OPENAI_DEPLOYMENT="gpt-4o"
export AZURE_OPENAI_API_VERSION="2025-01-01-preview"
# Launch the proxy server
node scripts/server.js
2. Update OpenClaw Configuration
Modify your ~/.openclaw/openclaw.json to point to the proxy:
{
"models": {
"providers": {
"azure-gpt4o": {
"baseUrl": "http://127.0.0.1:18790",
"apiKey": "YOUR_AZURE_API_KEY",
"api": "openai-completions",
"authHeader": false,
"headers": {
"api-key": "YOUR_AZURE_API_KEY"
},
"models": [{ "id": "gpt-4o", "name": "GPT-4o (Azure)" }]
}
}
}
}
Azure OpenAI Proxy Data Schema & Taxonomy
The proxy relies on environment variables for its configuration and exposes a simple health check schema:
| Variable | Purpose | Default |
|---|---|---|
| AZURE_PROXY_PORT | Local port for Openclaw Skills to connect to | 18790 |
| AZURE_OPENAI_ENDPOINT | Your specific Azure resource hostname | Required |
| AZURE_OPENAI_DEPLOYMENT | The name of your model deployment in Azure | gpt-4o |
Health Check Endpoint: GET /health returns {"status":"ok","deployment":"gpt-4o"}.
Azure OpenAI Proxy Advanced Features
- Background execution via systemd service support for persistent availability on Linux environments.
- Subagent specific routing to allow primary agents and subagents to use different Azure deployments.
- Support for custom API versions to maintain compatibility with Azure's preview features.
- Header override logic to satisfy Azure's unique api-key authentication requirements instead of standard Bearer tokens.
SKILL.md
Loading
Related Openclaw Skills
Lead Radar
Lead Radar is an automated lead generation agent that monitors social platforms for buying intent and delivers qualified prospects directly to Telegram.
Fiscal Personal Accountant
A sophisticated personal accountant skill that interfaces with the fscl CLI to manage Actual Budget environments through natural language.
Design Style Skill
A powerful design system integrator that enables AI agents to generate production-grade, aesthetically distinct frontend interfaces instead of generic UI.
Bilibili Downloader
A high-performance media extraction tool for Bilibili that supports 8K video downloads, audio extraction, and metadata retrieval.
Volcengine Configuration
A comprehensive onboarding and configuration guide to integrate Volcengine Ark coding models into your OpenClaw environment.
JSON2Video Pinterest
A powerful automation tool for creating professional, AI-powered vertical videos for Pinterest and social media.
Featured*
