This is an automated email from the ASF dual-hosted git repository. moonming pushed a commit to branch blog-manage-apisix-with-ai-agents in repository https://gitbox.apache.org/repos/asf/apisix-website.git
commit 0e3b5cd767dc16c4d46867ed2350361989fd13ed Author: Ming Wen <[email protected]> AuthorDate: Fri Jul 10 10:19:54 2026 +0800 blog: manage Apache APISIX with AI coding agents Add an educational blog post (en + zh) introducing open-source Agent Skills for Apache APISIX: describe an outcome in natural language and an AI coding agent runs the real a6 CLI commands against a live gateway. Links out to the full searchable catalog on docs.api7.ai. Frames a6 as an Apache-2.0 community CLI, not an official ASF release. --- .../07/10/manage-apache-apisix-with-ai-agents.md | 86 ++++++++++++++++++++++ .../07/10/manage-apache-apisix-with-ai-agents.md | 86 ++++++++++++++++++++++ 2 files changed, 172 insertions(+) diff --git a/blog/en/blog/2026/07/10/manage-apache-apisix-with-ai-agents.md b/blog/en/blog/2026/07/10/manage-apache-apisix-with-ai-agents.md new file mode 100644 index 00000000000..69c73385dad --- /dev/null +++ b/blog/en/blog/2026/07/10/manage-apache-apisix-with-ai-agents.md @@ -0,0 +1,86 @@ +--- +title: "Manage Apache APISIX with AI Coding Agents" +authors: + - name: "Ming Wen" + title: "Author" + url: "https://github.com/moonming"; + image_url: "https://github.com/moonming.png"; +keywords: + - Apache APISIX + - API Gateway + - AI Agent + - Agent Skills + - Claude Code + - AI Gateway +description: A walkthrough of managing Apache APISIX from your AI coding agent using open-source Agent Skills — describe the outcome you want, and the agent runs the real gateway commands. +tags: [Ecosystem] +--- + +Configuring an API gateway usually means remembering things: the exact Admin API path, the shape of a plugin's schema, which fields a route needs. For Apache APISIX, that knowledge is well-documented — but you still have to hold it in your head or keep a browser tab open while you work. + +AI coding agents change the ergonomics. What if, instead of looking up the `limit-count` schema, you could just say: *"Add key-auth to my `/orders` route and rate-limit it to 100 requests per minute"* — and your agent produced the correct configuration and applied it? + +That is exactly what **Agent Skills** for APISIX enable. + +<!--truncate--> + +## What are Agent Skills? + +An Agent Skill is a plain Markdown file — `SKILL.md` — with a bit of YAML frontmatter and a body of task-specific instructions. It teaches an AI coding agent how to accomplish one focused job: configure a specific plugin, run a specific operational recipe, or reason about a specific role. + +The format is deliberately simple and portable. Any agent that understands the Agent Skills convention can load them — including Claude Code, Cursor, GitHub Copilot, Windsurf, and OpenCode. There is no runtime, no plugin API, no lock-in: the skill is just knowledge the agent reads on demand. + +For Apache APISIX, these skills are published in the open-source [api7/a6](https://github.com/api7/a6) repository. `a6` is an Apache-2.0 licensed community command-line tool that wraps the APISIX Admin API; it is not part of an official Apache Software Foundation release. The skills teach an agent how to drive `a6` to configure a live gateway. + +## Describe the outcome, not the commands + +Here is the loop in practice. You tell your agent what you want: + +> *"Add key-auth to my `/orders` route and rate-limit it to 100 rpm."* + +The agent recognizes the intent, loads the `a6-plugin-key-auth` and `a6-plugin-limit-count` skills, generates the route configuration, and applies it with the file-based `a6` commands: + +```yaml +# route.yaml +uri: /orders +upstream_id: orders-upstream +plugins: + key-auth: {} + limit-count: + count: 100 + time_window: 60 + key: remote_addr + rejected_code: 429 +``` + +```bash +a6 route update orders -f route.yaml +a6 config sync +``` + +You reviewed a diff and confirmed an outcome; you did not look up a single schema field. + +## What's covered + +The skill set spans the full surface of day-to-day APISIX work, organized into four kinds: + +- **Plugins** — one skill per plugin, from `key-auth`, `jwt-auth`, and `limit-count` to `ai-proxy`, `prometheus`, and `zipkin`. Each covers configuration, examples, and common patterns. +- **Recipes** — multi-step operational tasks like canary releases, blue-green deployments, circuit breaking, health checks, and mTLS. +- **Personas** — role-oriented guidance for API developers and platform operators, with decision frameworks for choosing routes vs. services and which skills to load for a task. +- **Core** — shared conventions for working with the `a6` CLI. + +## Try it + +Skills are just files. Clone the repository and drop them into your agent's skills directory: + +```bash +git clone https://github.com/api7/a6.git +mkdir -p ~/.claude/skills +cp -r a6/skills/* ~/.claude/skills/ +``` + +Then describe what you want your gateway to do, in plain language. + +You can browse the full, searchable catalog — every plugin, recipe, and persona skill, with the commands each one runs — on the [AI Agent Skills page](https://docs.api7.ai/apisix/ai-agent-skills). The skills are Apache-2.0 licensed and open to community contributions in the [api7/a6](https://github.com/api7/a6) repository. + +AI agents are becoming a primary interface for infrastructure. Teaching them to speak APISIX fluently — in the open, with plain Markdown anyone can read and improve — is a natural next step for the project's ecosystem. diff --git a/blog/zh/blog/2026/07/10/manage-apache-apisix-with-ai-agents.md b/blog/zh/blog/2026/07/10/manage-apache-apisix-with-ai-agents.md new file mode 100644 index 00000000000..1e32a1ec4a7 --- /dev/null +++ b/blog/zh/blog/2026/07/10/manage-apache-apisix-with-ai-agents.md @@ -0,0 +1,86 @@ +--- +title: "用 AI 编码 Agent 管理 Apache APISIX" +authors: + - name: "Ming Wen" + title: "Author" + url: "https://github.com/moonming"; + image_url: "https://github.com/moonming.png"; +keywords: + - Apache APISIX + - API Gateway + - AI Agent + - Agent Skills + - Claude Code + - AI Gateway +description: 通过开源的 Agent Skills，从你的 AI 编码 Agent 管理 Apache APISIX——描述你想要的结果，Agent 就会执行真实的网关命令。 +tags: [Ecosystem] +--- + +配置 API 网关通常意味着要记住很多东西：Admin API 的确切路径、某个插件的 schema 结构、一条路由需要哪些字段。对 Apache APISIX 来说，这些知识都有完善的文档——但你仍然得把它记在脑子里，或者一边干活一边开着浏览器查。 + +AI 编码 Agent 改变了这种体验。如果你不用去查 `limit-count` 的 schema，而是直接说一句：*“给我的 `/orders` 路由加上 key-auth，并限流到每分钟 100 次请求”*——然后你的 Agent 就生成正确的配置并应用上去，会怎么样? + +这正是 APISIX 的 **Agent Skills** 所要实现的。 + +<!--truncate--> + +## 什么是 Agent Skills? + +一个 Agent Skill 就是一个纯 Markdown 文件——`SKILL.md`——带一小段 YAML frontmatter,以及一段针对具体任务的说明正文。它教会 AI 编码 Agent 完成一件聚焦的事:配置某个特定插件、执行某个特定的运维流程,或围绕某个特定角色进行决策。 + +这个格式刻意保持简单和可移植。任何理解 Agent Skills 约定的 Agent 都能加载它们——包括 Claude Code、Cursor、GitHub Copilot、Windsurf 和 OpenCode。没有运行时、没有插件 API、没有锁定:skill 只是 Agent 按需读取的知识。 + +针对 Apache APISIX,这些 skills 发布在开源的 [api7/a6](https://github.com/api7/a6) 仓库中。`a6` 是一个 Apache-2.0 许可的社区命令行工具,它封装了 APISIX 的 Admin API;它不是 Apache 软件基金会官方发行版的一部分。这些 skills 教会 Agent 如何驱动 `a6` 来配置一个运行中的网关。 + +## 描述结果,而不是命令 + +来看实际的循环。你告诉 Agent 你想要什么: + +> *“给 `/orders` 路由加上 key-auth,并限流到每分钟 100 次。”* + +Agent 识别出意图,加载 `a6-plugin-key-auth` 和 `a6-plugin-limit-count` 两个 skill,生成路由配置,并用基于文件的 `a6` 命令应用它: + +```yaml +# route.yaml +uri: /orders +upstream_id: orders-upstream +plugins: + key-auth: {} + limit-count: + count: 100 + time_window: 60 + key: remote_addr + rejected_code: 429 +``` + +```bash +a6 route update orders -f route.yaml +a6 config sync +``` + +你审查了一份 diff、确认了一个结果;而你没有去查任何一个 schema 字段。 + +## 覆盖了哪些内容 + +这套 skill 覆盖了日常 APISIX 工作的完整面,分为四类: + +- **Plugins(插件)**——每个插件一个 skill,从 `key-auth`、`jwt-auth`、`limit-count` 到 `ai-proxy`、`prometheus`、`zipkin`。每个都涵盖配置、示例和常见模式。 +- **Recipes(方案)**——多步运维任务,如金丝雀发布、蓝绿部署、熔断、健康检查和 mTLS。 +- **Personas(角色)**——面向 API 开发者和平台运维的角色化指引,包含选择路由还是服务、以及针对某个任务加载哪些 skill 的决策框架。 +- **Core(核心)**——使用 `a6` CLI 的共享约定。 + +## 试试看 + +Skills 就是文件。克隆仓库,把它们放进你 Agent 的 skills 目录即可: + +```bash +git clone https://github.com/api7/a6.git +mkdir -p ~/.claude/skills +cp -r a6/skills/* ~/.claude/skills/ +``` + +然后用自然语言描述你希望网关做什么。 + +你可以在 [AI Agent Skills 页面](https://docs.api7.ai/apisix/ai-agent-skills) 浏览完整的、可搜索的目录——每一个插件、方案和角色 skill,以及它们各自运行的命令。这些 skills 采用 Apache-2.0 许可,欢迎在 [api7/a6](https://github.com/api7/a6) 仓库中贡献。 + +AI Agent 正在成为基础设施的主要交互界面。用开放的方式——用任何人都能阅读和改进的纯 Markdown——教会它们流利地“说” APISIX,是这个项目生态自然的下一步。
![https://github.com/moonming.png"](https://github.com/moonming.png"){kind=link}