# AxonHub — Open-Source AI Gateway Platform

> AxonHub is an Apache-2.0 AI gateway with routing, failover, and trace viewing, letting you keep your SDK while centralizing keys, logs, and model config.

## Install

Save the content below to `.claude/skills/` or append to your `CLAUDE.md`:

## Quick Use

```bash
git clone https://github.com/looplj/axonhub.git && cd axonhub
export AXONHUB_DB_DSN='<USER>.root:<PASSWORD>@tcp(<HOST>:4000)/axonhub?tls=true&parseTime=true'
docker-compose up -d
docker-compose ps
```

## Intro

AxonHub is an Apache-2.0 AI gateway with routing, failover, and trace viewing, letting you keep your SDK while centralizing keys, logs, and model config.

**Best for:** teams that need one gateway to standardize model access, keys, routing, and observability

**Works with:** existing LLM SDKs, Docker Compose, optional Helm/Kubernetes deployments, SQL backends for state

**Setup time:** 15-25 minutes

### Key facts (verified)

- GitHub: 3721 stars · 443 forks · pushed 2026-05-12.
- License: Apache-2.0 · owner avatar + repo URL verified via GitHub API.
- README-verified entrypoint: `git clone https://github.com/looplj/axonhub.git && cd axonhub`.

## Main

- Keep app code stable: point existing SDKs at the gateway, then evolve routing and key management centrally.

- Use traces and request monitoring to debug agent failures as distributed systems problems (latency, retries, model swaps).

- Start with a single environment (dev) and a small model set; expand only after you have a clean policy for keys and logging.

### Source-backed notes

- README provides a quick install using a release tarball and an `./axonhub` binary.
- README includes docker-compose startup instructions and an `axonhub config check` command.
- LICENSE overview states most of the repo is Apache-2.0, with some scoped subdirectory licensing notes.

### FAQ

- **Do I need Kubernetes?**: No. README includes docker-compose and binary install paths.
- **What’s the main value?**: Central routing + observability without rewriting every client integration.
- **Is it fully Apache-2.0?**: Mostly; README/license notes mention scoped subdirectory licenses you should review.

## Source & Thanks

> Source: https://github.com/looplj/axonhub
> License: Apache-2.0
> GitHub stars: 3721 · forks: 443

---

<!-- ZH -->

## Quick Use

```bash
git clone https://github.com/looplj/axonhub.git && cd axonhub
export AXONHUB_DB_DSN='<USER>.root:<PASSWORD>@tcp(<HOST>:4000)/axonhub?tls=true&parseTime=true'
docker-compose up -d
docker-compose ps
```

## Intro

AxonHub 是 Apache-2.0 许可的开源 AI 网关与观测平台，提供路由/故障切换、请求监控与 trace 查看，让你保留现有 SDK 的同时集中管理密钥、日志与模型配置，并支持本地或容器化部署。

**Best for:** 需要统一模型访问、密钥、路由与观测的一体化网关团队

**Works with:** 现有 LLM SDK、Docker Compose、可选 Helm/K8s 部署、SQL 后端存储状态

**Setup time:** 15-25 minutes

### Key facts (verified)

- GitHub：3721 stars · 443 forks；最近更新 2026-05-12。
- 许可证：Apache-2.0；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中核对过的入口命令：`git clone https://github.com/looplj/axonhub.git && cd axonhub`。

## Main

- 保持应用代码不动：让现有 SDK 指向网关，再在中心化层面演进路由与密钥策略。

- 用 trace 与请求监控把 agent 失败当分布式系统问题来定位（延迟、重试、模型切换）。

- 先从单一环境（dev）与少量模型开始；在密钥与日志策略稳定后再扩大范围。

### Source-backed notes

- README 提供了通过 release 包解压后运行 `./axonhub` 的安装方式。
- README 包含 docker-compose 启动流程，并提供 `axonhub config check` 配置校验命令。
- LICENSE 概览说明主体为 Apache-2.0，部分子目录有单独的许可说明。

### FAQ

- **一定要上 Kubernetes 吗？**：不需要。README 同时给出 docker-compose 与二进制安装方式。
- **核心价值是什么？**：在不重写各个客户端集成的前提下，集中实现路由与可观测性。
- **完全是 Apache-2.0 吗？**：主体是；但许可说明里提到部分子目录有额外规则，建议按需复核。

## Source & Thanks

> Source: https://github.com/looplj/axonhub
> License: Apache-2.0
> GitHub stars: 3721 · forks: 443


---
Source: https://tokrepo.com/en/workflows/axonhub-open-source-ai-gateway-platform
Author: AI Open Source