lcp-openai-serve

Develop an OpenAI-compatible HTTP gateway in Go with strict logging controls.

2|Updated Dec 30, 2025
One-click install
npx skills add https://github.com/YusukeShimizu/lightning-compute-protocol --skill lcp-openai-serve
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lcp-openai-serve
Source: https://github.com/YusukeShimizu/lightning-compute-protocol/tree/main/.codex/skills/lcp-openai-serve
Command: npx skills add https://github.com/YusukeShimizu/lightning-compute-protocol --skill lcp-openai-serve

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a development blueprint for building an OpenAI-compatible HTTP gateway in Go, with a focus on prompt privacy and robust module boundaries.

Core Features & Use Cases

  • Modular Gateway: A focused apps/openai-serve module that implements a strict, SRP-aligned gateway interface.
  • Security & Logging: Logging controls ensure prompts and outputs are not logged, with configurable verbosity.
  • Workflow & Validation: Clear developer workflow that emphasizes tests, linting, and strict decoding of inputs to prevent unsafe behavior.

Quick Start

Read apps/openai-serve/README.md before changing behavior. Run make test, make lint, and make fmt from the repository root to validate changes.

Frequently Asked Questions about lcp-openai-serve

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build an OpenAI-compatible HTTP gateway in Go that preserves prompt privacy?

An OpenAI-compatible HTTP gateway in Go preserves prompt privacy by enforcing strict logging controls that prevent prompts and outputs from being logged, while maintaining configurable verbosity for operational visibility.

What is the best way to enforce module boundaries in a Go gateway?

Enforcing module boundaries in a Go gateway involves applying SRP-aligned design within a focused module, ensuring strict JSON decoding of inputs, and preventing unsafe behavior through isolated component responsibilities.

Do I need specific testing and linting tools to validate a Go OpenAI gateway?

Validating a Go OpenAI gateway requires running make test, make lint, and make fmt from the repository root to ensure code quality, formatting compliance, and behavioral correctness before deployment.

How do I configure environment-driven settings for an OpenAI-compatible gateway?

Environment-driven configuration for an OpenAI-compatible gateway is implemented by reading environment variables to set operational parameters, with developer guidelines documented in the module's README for setup instructions.

Can I use strict JSON decoding to prevent unsafe behavior in a Go HTTP gateway?

Strict JSON decoding prevents unsafe behavior in a Go HTTP gateway by validating and sanitizing all incoming request payloads, rejecting malformed inputs before they reach the core gateway processing logic.

Why are prompts not being logged in my OpenAI gateway implementation?

Prompts are not logged because the gateway enforces strict logging controls designed to protect prompt privacy by default, ensuring sensitive inputs and outputs are excluded from application logs.