<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
    <id>https://blog.ax0x.ai/</id>
    <title>A Cat Entertainer</title>
    <updated>2026-05-21T16:22:32.929Z</updated>
    <generator>https://github.com/jpmonette/feed</generator>
    <author>
        <name>Xingfan Xia</name>
        <email>xingfanxia@gmail.com</email>
        <uri>https://blog.ax0x.ai</uri>
    </author>
    <link rel="alternate" href="https://blog.ax0x.ai/"/>
    <subtitle>Just A Tech Blog</subtitle>
    <icon>https://blog.ax0x.ai/favicon.svg</icon>
    <rights>All rights reserved 2026, Xingfan Xia</rights>
    <entry>
        <title type="html"><![CDATA[我把养了半年的 Claude Code 配置开源了]]></title>
        <id>https://blog.ax0x.ai/dot-claude-public-zh</id>
        <link href="https://blog.ax0x.ai/dot-claude-public-zh"/>
        <updated>2026-05-09T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[50+ skill、50+ subagent、40+ slash command 的完整 Claude Code 配置，加上一套 fail-closed 双层审计的 sync 基础设施，整套推到了 GitHub 公开仓库。这篇聊里面有什么、为什么这么搞、以及那套自己摸出来的工作流哲学。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[I Open-Sourced My Entire ~/.claude]]></title>
        <id>https://blog.ax0x.ai/dot-claude-public</id>
        <link href="https://blog.ax0x.ai/dot-claude-public"/>
        <updated>2026-05-09T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Six months of accumulated Claude Code configuration — 50 skills, 50 subagents, 40 commands, hooks, rules, and the sync pipeline that pushes all of it to a public GitHub mirror with a fail-closed two-tier audit. What's inside, why I built it, and the workflow philosophy behind it.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[攒了一年的 11 个 AI skill，开源了]]></title>
        <id>https://blog.ax0x.ai/ax-skills-zh</id>
        <link href="https://blog.ax0x.ai/ax-skills-zh"/>
        <updated>2026-05-08T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[AX-skills 是我自己每天在用的 Claude Code skill 仓库。写了一年多，写着写着发现 skill 不只是脚本，是把模糊的工作流明确化、让 AI 能继承「关心」的方式。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[I Open-Sourced the 11 AI Skills I Use Every Week]]></title>
        <id>https://blog.ax0x.ai/ax-skills</id>
        <link href="https://blog.ax0x.ai/ax-skills"/>
        <updated>2026-05-08T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[AX-skills is the public mirror of my personal Claude Code skills directory — eleven workflows for writing, research, image generation, knowledge sync, and one cyber half-immortal divination skill. After a year of writing them, I think a skill isn't a stored prompt — it's how you encode care.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[朋友试了我推荐的 8 个会议录音工具，全部退货]]></title>
        <id>https://blog.ax0x.ai/meeting-recorder-oss-survey-v2-zh</id>
        <link href="https://blog.ax0x.ai/meeting-recorder-oss-survey-v2-zh"/>
        <updated>2026-05-08T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[上一篇翻了 8 个开源会议录音工具，朋友试完反馈三个维度全垃圾——识别烂、总结烂、UI/UX 烂。我懵了，重新挖了一晚上。这次发现根因比想的深：所有这些工具底层都是同一代 Whisper，而 Whisper 这代技术栈早就被超车了。这篇讲数据、讲被维护者自己都认账的 issue、讲一个被所有人忽视的暴力解，还有上次漏掉的几个真正值得试的项目。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[I Recommended Eight Open-Source Meeting Recorders. My Friend Returned All of Them.]]></title>
        <id>https://blog.ax0x.ai/meeting-recorder-oss-survey-v2</id>
        <link href="https://blog.ax0x.ai/meeting-recorder-oss-survey-v2"/>
        <updated>2026-05-08T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Last week I surveyed 8 open-source meeting recorders. My friend tried every one. Verdict: ASR is bad, summaries are bad, UI/UX is bad. So I dug deeper. The real problem is structural — every project I recommended is built on Whisper, and Whisper has been outpaced by 2-4× on Chinese benchmarks for over a year. This is the deeper survey: data, the issues that maintainers themselves admit, the dumb-but-effective hack everyone overlooked, and four projects I missed the first time.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[Zoom 自带的 AI 太烂，我把开源会议录音工具翻了个遍]]></title>
        <id>https://blog.ax0x.ai/meeting-recorder-oss-survey-zh</id>
        <link href="https://blog.ax0x.ai/meeting-recorder-oss-survey-zh"/>
        <updated>2026-05-08T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[前两天有朋友问我有没有好用的会议录音工具——Zoom/Teams 自带的 AI 总结烂得没法看，Granola/Krisp 都要订阅，他用 Gsuite 但不开 Meet。我翻了一晚上 GitHub，从 11.6k 星的 Meetily 到三五个 star 的玩具，把整个开源星座摊开来。这篇讲谁是产品、谁是玩具、谁适合哪种人。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[Zoom's Native AI Sucks. So I Surveyed Every Open-Source Meeting Recorder]]></title>
        <id>https://blog.ax0x.ai/meeting-recorder-oss-survey</id>
        <link href="https://blog.ax0x.ai/meeting-recorder-oss-survey"/>
        <updated>2026-05-08T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[A friend asked for a meeting recorder. Zoom/Teams native AI is unusable. Granola and Krisp work but cost ~$200/yr per seat. He's on Google Workspace but doesn't use Meet, so the bundled Gemini summarizer is dead weight. I spent a night on GitHub and found a whole constellation — from Meetily at 11.6k stars down to single-digit-star personal toys, all racing to commoditize the same three Lego pieces.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[为了打 AI Agent 邪修赛道，我把算命产品拆了]]></title>
        <id>https://blog.ax0x.ai/skill-workshop-01-banxian-zh</id>
        <link href="https://blog.ax0x.ai/skill-workshop-01-banxian-zh"/>
        <updated>2026-05-06T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[飞书办了个 Skill 比赛，有个赛道叫邪修。我把 29 天造的算命产品里三个东方占卜模块抠出来，打包成一个赛博半仙 skill。这一篇是过程记录，也是 AX Skill Workshop 系列开篇。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[Packaging a Product into a Skill: A Cyber-Banxian for an Agent Contest]]></title>
        <id>https://blog.ax0x.ai/skill-workshop-01-banxian</id>
        <link href="https://blog.ax0x.ai/skill-workshop-01-banxian"/>
        <updated>2026-05-06T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Took the three Eastern divination engines from PanPanMao and packaged them into a single agent-portable skill — algorithm engine + knowledge base + persona prompt, three folders deep. The pattern transfers to any product with real domain knowledge.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[从万物回到三个品类]]></title>
        <id>https://blog.ax0x.ai/shichuan-per-domain-templates-zh</id>
        <link href="https://blog.ax0x.ai/shichuan-per-domain-templates-zh"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Part 2 的「6 槽位通吃 7 品类」是一层漂亮的抽象。跑了两周，测出牛仔外套生成「宝石星图」的 bug。这一篇讲抽象失守在哪一层、为什么没用户时分阶段上线是假安全感、以及把服饰和美妆各自推倒重来的过程。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[Walking Back the Abstraction]]></title>
        <id>https://blog.ax0x.ai/shichuan-per-domain-templates</id>
        <link href="https://blog.ax0x.ai/shichuan-per-domain-templates"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Part 2 claimed six slots could cover seven categories. Two weeks later, a denim jacket generated a gem constellation. This post walks through where the abstraction failed, why skipping the 'safe' phased rollout was the right call, and how each of three categories got its own six-template set.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[一下午让 AI 出六个设计方向，solo dev 的设计终于不用将就]]></title>
        <id>https://blog.ax0x.ai/claude-design-pipeline-zh</id>
        <link href="https://blog.ax0x.ai/claude-design-pipeline-zh"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Solo dev 做设计，大多数时候 N=1——第一直觉就是唯一方案，因为做第二个版本要花一天。Claude Design 把这事拆了：一下午六个完全不同的视觉语言摆在面前。留下了什么、丢了什么、方向锁死之后 big-task 的 ui profile 接手怎么走——这篇讲这些。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[Six Directions in an Afternoon: Using Claude Design to Kill First-Instinct Bias]]></title>
        <id>https://blog.ax0x.ai/claude-design-pipeline</id>
        <link href="https://blog.ax0x.ai/claude-design-pipeline"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[For solo dev and small teams, design exploration is usually N=1 — you go with your first instinct because producing a second variant takes a day. Claude Design collapses that: six distinct aesthetic directions in an afternoon. What I kept from the pipeline, what I dropped, and how `big-task`'s UI profile takes over once a direction is locked.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[两个月，我把 AI 工作流推倒重来了四次]]></title>
        <id>https://blog.ax0x.ai/harness-iteration-zh</id>
        <link href="https://blog.ax0x.ai/harness-iteration-zh"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[先是裸跑 Claude Code。再是 GSD 多 agent 仪式。再是 superpowers 的 skill 库。最后在上面包了一个自己的 orchestrator。每一版崩盘都来自一个具体 session。最后我才想明白：solo dev 的 harness 跟团队的不是一个物种，而且真正的答案不是选一个默认引擎——是按工作形状路由。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[I Rewrote My AI Coding Harness Four Times in Two Months]]></title>
        <id>https://blog.ax0x.ai/harness-iteration</id>
        <link href="https://blog.ax0x.ai/harness-iteration"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Vanilla Claude Code. Then GSD's multi-agent ceremony. Then superpowers' skill system. Then a custom orchestrator on top. Each rewrite came from a specific session that broke. What I learned: solo dev harnesses and team harnesses are different species, and the right answer isn't a default engine — it's routing by work shape.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[同一套 AI 设计流水线，三个产品跑出三种形状]]></title>
        <id>https://blog.ax0x.ai/three-product-shapes-zh</id>
        <link href="https://blog.ax0x.ai/three-product-shapes-zh"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[同一条流水线——Claude Design 出方向、HANDOFF.md 锁死、`big-task` 落地——但每个产品都把它掰成不一样的形状。ÉLAN 是空地，需要宽探索。识川 app 已经上线，要按类目一个个迁移。盘盘猫是 monorepo，v2 跑三个并行品牌表达而不是一个方向。变的是什么，不变的是什么。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[One Pipeline, Three Shapes: Running Claude Design + big-task Across ÉLAN, Shichuan, and Panpanmao]]></title>
        <id>https://blog.ax0x.ai/three-product-shapes</id>
        <link href="https://blog.ax0x.ai/three-product-shapes"/>
        <updated>2026-04-23T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[Same pipeline — Claude Design explores, HANDOFF.md locks, `big-task` ships — but it bends around each product. ÉLAN is greenfield and needs wide exploration before converging. Shichuan has a live app and needs incremental category-by-category migration. Panpanmao is a monorepo with a v2 redesign that runs three parallel brand expressions instead of one. What changes and what doesn't.]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[每天 488 条 AI 新闻，我让一份 Markdown 替我选题]]></title>
        <id>https://blog.ax0x.ai/radar-01-policy-as-markdown-zh</id>
        <link href="https://blog.ax0x.ai/radar-01-policy-as-markdown-zh"/>
        <updated>2026-04-20T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[AX 的 AI 雷达每天从 50 多个信源抽约 500 条内容，挑 15 条给我。选题不靠 prompt，靠一份能版本化、能 diff 的 Markdown 编辑策略——一个 Agent 从反馈里迭代它。讲这把雷达的架构取舍和关键设计决策。]]></summary>
    </entry>
    <entry>
        <title type="html"><![CDATA[488 AI Stories a Day: Why My Editorial Policy Is a Markdown File, Not a Prompt]]></title>
        <id>https://blog.ax0x.ai/radar-01-policy-as-markdown</id>
        <link href="https://blog.ax0x.ai/radar-01-policy-as-markdown"/>
        <updated>2026-04-20T00:00:00.000Z</updated>
        <summary type="html"><![CDATA[AX Radar filters ~500 stories a day from 50+ sources down to about 15. Editorial judgment isn't a prompt — it's a versionable, diff-able Markdown file that a Claude Agent iterates against user feedback. A breakdown of the architecture and the design calls.]]></summary>
    </entry>
</feed>