claude-worker-proxy

🥈Silver

Cloudflare Worker proxy that converts API calls from Gemini and OpenAI models to Claude Code format. Enables low-cost, zero-configuration access to Claude Code with One-Balance integration. Supports streaming and non-streaming responses, tool calls, and easy deployment.

2681240Updated Today

Intermediate30min to implementautomation

Saves ~60 min per use

Quick InstallView Source

git clone https://github.com/glidea/claude-worker-proxy.git

Works with:

Claude

Overview

About This Skill

How to Use

[{"step":"Prepare your API call. Write your prompt in the format expected by OpenAI or Gemini models, including any required tools or parameters. For example: 'Analyze customer feedback data in [DATASET] and classify sentiment using [MODEL]. Use [TOOLS] for execution.'","tip":"Use the Claude Code documentation to ensure your prompt includes all necessary components like tools, streaming flags, and message structure."},{"step":"Deploy the Cloudflare Worker proxy. Clone the proxy repository from [GITHUB_REPO] and run `npm install` followed by `wrangler deploy --name claude-worker-proxy`. Configure your One-Balance API key in the worker's environment variables.","tip":"Use the Wrangler CLI for deployment. Ensure your Cloudflare account has Workers enabled and sufficient quota for your expected usage volume."},{"step":"Convert your API call. Send your OpenAI/Gemini-style request to the deployed worker endpoint. The proxy will automatically convert the request format to match Claude Code's requirements. Use curl for testing: `curl -X POST https://claude-worker-proxy.your-subdomain.workers.dev -H 'Content-Type: application/json' -d '{\"messages\": [...]}'`.","tip":"Enable streaming by setting 'stream': true in your request body. This reduces latency for large responses."},{"step":"Process the response. The proxy will return a Claude Code-compatible response with tool calls, text chunks, or final output. Parse the response to extract results or handle tool executions as needed.","tip":"For tool calls, implement a handler to execute the code in a sandboxed environment (e.g., using Docker or a cloud function) and return the results to the proxy."},{"step":"Monitor usage and costs. Check your One-Balance dashboard for token usage, costs, and performance metrics. Adjust your deployment settings (e.g., concurrency limits) based on observed traffic patterns.","tip":"Set up Cloudflare Workers Analytics to track request volumes, latency, and errors. Use this data to optimize your proxy configuration."}]

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/glidea/claude-worker-proxy

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Use the Claude Worker Proxy to convert [OPENAI_API_CALL] or [GEMINI_API_CALL] into a Claude Code-compatible format. Deploy the proxy to Cloudflare Workers using [DEPLOYMENT_COMMAND]. Ensure the proxy is configured with your One-Balance API key. Test the converted call with [TEST_INPUT] and verify the output matches expected Claude Code behavior.

Example Output

```json
{
  "converted_request": {
    "model": "claude-3-5-sonnet-20241022",
    "messages": [
      {
        "role": "user",
        "content": "Write a Python script to analyze a CSV file with columns 'date', 'revenue', and 'expenses'. Calculate monthly profit and generate a line chart using matplotlib."
      }
    ],
    "tools": [
      {
        "name": "code_interpreter",
        "description": "A tool that executes code in a sandboxed environment and returns the output",
        "input_schema": {
          "type": "object",
          "properties": {
            "code": {
              "type": "string",
              "description": "The Python code to execute"
            }
          },
          "required": ["code"]
        }
      }
    ],
    "stream": true
  },
  "proxy_response": {
    "status": "success",
    "response_id": "msg_1234567890",
    "content": [
      {
        "type": "text",
        "text": "Here's a Python script to analyze your CSV file and generate the requested visualization..."
      },
      {
        "type": "tool_call",
        "name": "code_interpreter",
        "arguments": {
          "code": "import pandas as pd\nimport matplotlib.pyplot as plt\n\ndf = pd.read_csv('data.csv')\ndf['profit'] = df['revenue'] - df['expenses']\ndf['date'] = pd.to_datetime(df['date'])\nmonthly_profit = df.groupby(df['date'].dt.to_period('M'))['profit'].sum()\n\nplt.figure(figsize=(10, 6))\nmonthly_profit.plot(kind='line', marker='o')\nplt.title('Monthly Profit')\nplt.xlabel('Date')\nplt.ylabel('Profit ($)')\nplt.grid(True)\nplt.savefig('monthly_profit.png')\nplt.close()"
        }
      }
    ],
    "usage": {
      "input_tokens": 125,
      "output_tokens": 450,
      "total_tokens": 575
    }
  }
}
```

The proxy successfully converted the OpenAI-style API call into a Claude Code-compatible request, deployed the Cloudflare Worker, and returned a streaming response with tool calls. The generated script was executed in a sandboxed environment, producing the expected line chart saved as 'monthly_profit.png'. The total cost was $0.0023, demonstrating the low-cost advantage of the One-Balance integration.