Back to feed
Dev.to
Dev.to
7/17/2026
How I built an OpenAI-compatible prompt encryption proxy in 300 lines of Python

How I built an OpenAI-compatible prompt encryption proxy in 300 lines of Python

Short summary

A developer who sells GPT prompts built a 300-line Python proxy using FastAPI and httpx that injects system prompts server-side, so buyers never see the prompt text. The proxy is OpenAI-compatible with streaming SSE support, bearer key auth, and works with clients like ChatBox and Cursor. The open-source Lite version is on GitHub; a paid Pro version adds encrypted storage and multi-tenant support.

  • •300-line FastAPI proxy injects system prompts server-side, preventing prompt theft
  • •OpenAI-compatible API with streaming SSE, bearer auth — buyers point their client at your URL
  • •Mistakes and fixes covered: SSE format re-emission, system prompt prepending, timeout tuning

Generated with AI, which can make mistakes.

#ai-tools#open-source
Read full article at Dev.to

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

The original title is "I Built a Privacy Scrubber So My Prompts Stop Leaking Data to AI Models"
Dev.to
I Built a Prompt Compressor That Saves 65% on LLM Costs — Here's the Story
Dev.to
The original title is: "CachePilot: A drop-in caching proxy for OpenAI/Anthropic/OpenRouter APIs"
Dev.to
The original title is "How I Cut LLM API Costs by 60% With 2 Lines of Code"
Dev.to