dyndynjyxa/aio-coding-hub

一个All In One的本地AI工具, 支持Win/Mac/Linux

What it solves

AIO Coding Hub provides a unified local gateway for AI coding CLI tools (such as Claude Code, Codex, and Gemini CLI). It eliminates the need to configure API keys for each tool individually, prevents request failures through intelligent failover, and provides visibility into token usage and costs across different providers.

How it works

The application acts as a local proxy (listening on 127.0.0.1:37123) that intercepts requests from coding CLIs. It routes these requests to various AI providers based on user-defined priority lists (sorting templates). If a primary provider fails, the gateway automatically switches to a backup provider using a circuit breaker pattern to ensure stability.

Who it’s for

Developers who use multiple AI coding CLI tools and multiple LLM providers, and who want a centralized way to manage their API configurations, monitor costs, and ensure high availability of their AI-assisted coding workflow.

Highlights

  • Unified Gateway: Single entry point for multiple coding CLIs with independent on/off switches.
  • Intelligent Routing: Automatic failover and circuit breaker protection to handle unstable upstream providers.
  • Smart Observability: Detailed token usage statistics, cost estimation, and request trace logs with heatmaps.
  • Workspace Management: Project-specific isolation for Prompts, MCP, and Skill configurations.
  • Extensible Plugin System: Support for custom commands, provider extensions, and gateway hooks via an Extension Host.
  • Model Validation: Tools to verify provider capabilities like token truncation and extended thinking across different vendors.

Related

  • Project
    musistudio/claude-code-router

    Claude Code Router (CCR) is a local model gateway that unifies many coding‑focused AI agents (Claude Code, Codex, Grok, Kimi, etc.) behind a single endpoint. It lets you manage providers, API keys, routing rules, retries, and tool extensions from a desktop UI, CLI, or Docker container, and provides detailed logs and cost estimates for every request.

  • Project
    james-6-23/codex2api

    Codex2API is a self‑hosted, Docker‑ready gateway that aggregates multiple Codex (OpenAI‑style) accounts into a single OpenAI/Anthropic‑compatible API. It handles token refresh, health‑based scheduling, rate‑limit recovery, per‑account billing, and provides a React admin dashboard for token import, API‑key management, usage analytics, and image/video generation.

  • Project
    jazzenchen/VibeAround

    VibeAround is a desktop/CLI hub that lets you launch many AI coding agents (Claude, Codex, Gemini, etc.) with any supported model provider via a local API‑shape bridge, offers live request inspection, session hand‑off to chat apps, web‑based terminal and dashboard, remote tunnelling, live preview, and host‑side web search—all while keeping execution on your own computer.

  • Project
    awslabs/aidlc-workflows

    AI‑DLC Workflows (awslabs/aidlc‑workflows) supply rule files that turn popular AI coding assistants (Kiro, Amazon Q, Cursor, Cline, Claude Code, GitHub Copilot, OpenAI Codex) into a structured, self‑correcting development workflow. After downloading the release zip, you copy the `aws-aidlc-rules` and `aws-aidlc-rule-details` folders into the hidden configuration directory required by your IDE/agent. The core workflow (`core‑workflow.md`) defines a three‑phase adaptive process (inception → construction → operations) that the AI follows, prompting you for confirmations and automatically handling corrections. The project is a genuine AWS‑maintained tool for AI‑driven software development, licensed under Apache‑2.0.

  • Project
    ding113/claude-code-hub

    An AI API proxy and management platform that provides unified access, load balancing, and cost tracking for multiple AI providers like Claude and OpenAI.