korallis

Ez Prox — AI skill for Claude Code

AI community

One-command macOS installer to route AI subscriptions (Claude Max, GitHub Copilot, OpenAI Max, Antigravity) through a local proxy into Factory Droid CLI with max reasoning enabled.

How to install Ez Prox

This entry records only its repository, not the path inside it, so there is no exact command to give. Open korallis/Ez-Prox and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Ez Prox does

One-command macOS installer to route AI subscriptions (Claude Max, GitHub Copilot, OpenAI Max, Antigravity) through a local proxy into Factory Droid CLI with max reasoning enabled.

Alternatives in AI

UI UX Pro Max Skill — An AI SKILL that provide design intelligence for building professional UI/UX multiple platforms 45.8k ★
Graphify — AI coding assistant skill (Claude Code, Codex, OpenCode, OpenClaw, Factory Droid, Trae) 20.9k ★
Vibeproxy — Native macOS menu bar app to use your Claude Code & ChatGPT subscriptions with AI coding tools - no API keys n 3.3k ★

README

Ez-Prox

**One-command setup to route your existing AI subscriptions through a local proxy into Factory Droid CLI — with max reasoning forced on every model.**

Ez-Prox is an interactive macOS installer that configures [CLIProxyAPIPlus](https://github.com/router-for-me/CLIProxyAPIPlus) as a local proxy server, wires it into [Factory Droid CLI](https://docs.factory.ai), and injects a shell function so you can launch everything with a single command: `droidc`.

What It Does

Instead of paying for separate API keys, Ez-Prox lets you funnel your **existing AI subscriptions** through a local proxy on port `8317`. The proxy handles OAuth authentication and presents all your models to Droid CLI as if they were standard API endpoints.

Supported Providers & Models

Provider	Models	Auth Method
Claude Max	Opus 4.6 (1M context), Sonnet 4.6, Haiku 4.5	OAuth (browser)
GitHub Copilot	GPT-5.4, GPT-5.4 Mini, GPT-5.3-Codex, Gemini 3.1 Pro	OAuth (browser)
OpenAI Max / Codex	GPT-5-Codex, GPT-5	OAuth (browser)
Antigravity	Gemini 3 Pro, Gemini 3 Flash	OAuth (browser)

Max Reasoning — Forced at the Proxy Level

The proxy injects reasoning/thinking parameters into **every request** before it reaches the upstream provider:

Claude models: thinking.type: enabled with a 30,000-token thinking budget and 128K max output
GPT models (excluding mini): reasoning.effort: high
Gemini models: generationConfig.thinkingConfig.thinkingLevel: high

GPT-5.4 Mini is intentionally left at default reasoning — it is used as a fast validation worker in Droid missions.

How It Works

┌──────────┐      ┌───────────────────────┐      ┌──────────────────┐
│ Droid CLI │─────▶│ CLIProxyAPIPlus :8317 │─────▶│ Claude / OpenAI  │
│           │      │  • Injects reasoning  │      │ / Copilot /      │
│           │◀─────│  • OAuth token mgmt   │◀─────│ Antigravity      │
└──────────┘      └─────────