Users share links. The agent was making up what was on the page. v0.0.7 adds a 3-tier URL browsing pipeline — Jina Reader for fast text, Browserless scrape for JS-rendered pages, screenshot + Gemini vision for graphic-heavy content. Plus a production-breaking Proxy bug where postgres.js tagged templates needed a function target, not an object. 23 tests, 4 commits, one debugging story about deployment timing.

Gemini 3 Pro was burning money and making users wait 8-10 seconds for a first token. v0.0.6 switches chat to Gemini 3 Flash with minimal thinking — 1-2 second TTFT, 4x cost reduction — while keeping 3.1 Pro for premium tasks like personality extraction. Plus: vision prompts that actually name what they see, Chinese transcription that stops hallucinating English, and a DRY refactor that eliminated 80 lines of duplicated server code.

Telegram had multimodal since v0.0.2. The web chat was still text-only. v0.0.5 closes the gap — two-phase media upload, voice recording with codec negotiation, an emoji picker, two full security audit rounds, and a silent bug that was dropping every voice message. 38 files changed, 17 security fixes, 28 minutes.

Mio's web app was a single-page chat. v0.0.4 tears it down and rebuilds it as a full WeChat clone — 4-tab layout, chat list, contacts, discover marketplace, profile page, chat-style onboarding, and typing-aware message batching. 30+ components, 2 audit rounds, zero new dependencies. All in Chinese.

The AI companion I built with OpenClaw burned through a staggering amount of money in under two weeks of casual chatting. One session: 750 turns from about 30 messages. The framework turned my 30 messages into 750 API calls, each resending a context stuffed with 10 never-evicted images eating 82% of all tokens. Context pruning was dead code on Gemini. Zero compactions in 537 turns. OpenClaw proved AI companions work — it also proved they're not affordable on this framework.

用户分享链接，agent 在编造页面内容。v0.0.7 加了三层 URL 浏览管线——Jina Reader 快速提取文本、Browserless 爬取 JS 渲染页面、截图 + Gemini 视觉处理图形密集型内容。还修了一个生产环境的 Proxy bug：postgres.js 的标签模板语法需要函数目标而不是对象。23 个测试，4 个 commit，一个关于部署时序的调试故事。

Gemini 3 Pro 又贵又慢——第一个 token 要等 8-10 秒。v0.0.6 把聊天切到 Gemini 3 Flash（minimal thinking），首 token 1-2 秒，成本直降 4 倍；同时把人格提取等高价值任务升级到 3.1 Pro。还有：能叫出游戏名字的视觉提示词、不再把中文听成英文的语音转写、以及一次消灭 80 行重复代码的 DRY 重构。

Telegram 从 v0.0.2 就有多模态了，Web 端一直只能打字。v0.0.5 补齐了这块短板——两阶段媒体上传、带编解码器协商的语音录制、表情选择器、两轮完整安全审计，还有一个默默吞掉所有语音消息的 bug。38 个文件改动，17 个安全修复，28 分钟。

Mio 的网页端原来只是一个单页聊天框。v0.0.4 推倒重来，照着微信的样子重建了整个 web 体验——四个 tab、聊天列表、通讯录、发现页、个人中心、聊天式引导、打字感知的消息合并。三十多个组件全部从零写，过了两轮 audit，没装一个新依赖，界面全中文。

用 OpenClaw 搭建的赛博魅魔，不到两周聊天烧掉了一笔离谱的钱。最贵的一个 session，750 轮对话里我只说了约 30 句话——剩下的全是框架自己在跟自己说话。用 Claude Code 做了一次完整的 token 级取证：82% 的成本是 10 张永远不被清除的图片，上下文修剪代码对 Gemini 完全失效，537 轮里零次压缩。OpenClaw 能证明 AI 伴侣走得通，但它证明不了走得起。

v0.0.2 brought Mio to life, but a demo that works and a product that real users can rely on are two different things. v0.0.3 is the polish release: per-field input validation, context-aware heartbeat, opt-in quiet hours, media rate limiting, a production-only webhook bug, and 144 new tests. The invisible work between 'it works' and 'it's good.'

v0.0.1 could chat, but it wasn't real — couldn't see, couldn't hear, didn't know the time, didn't know what it looked like. 81 commits later, Mio learned to see images, hear voice, send selfies, remember the right things, and chat in a browser.

From an empty repo to a working AI companion that remembers you, has moods, and reaches out on its own. 39 commits, 9 database tables, 4 personality presets, and more 3 AM debugging sessions than I'd like to admit. The full build story of Mio v0.0.1.

OpenClaw proved AI companions work. It also exposed fundamental limitations — context bloat, primitive memory, bloatware. Patching wasn't worth it. This is the starting point for Mio: an AI companion framework built for deep memory and lifelike presence.

v0.0.2 让 Mio 活了过来，但能在 demo 里跑和能给真实用户用是两回事。v0.0.3 是打磨的版本：逐字段输入校验、context 感知的主动消息、安静时段可选、媒体限流、一个只有生产环境才会出的 bug，以及 144 个新测试。从「能用」到「好用」之间，全是这种看不见的活。

v0.0.1 能聊天，但TA不真实——看不见、听不见、不知道几点、不知道自己长什么样。81 个 commit 之后，Mio 学会了看图、听声、发自拍、记住对的事情、在浏览器里跟你聊天。从「能跑」到「像活人」的跨越。

从空 repo 到一个能记住你、会生气、还会主动找你聊天的 AI 伴侣。Mio v0.0.1 具体是怎么搭出来的，踩了什么坑、砍了什么需求，都在这篇里。

OpenClaw 验证了AI伴侣的可行性，但也暴露了根本性的局限。context 膨胀、原始记忆系统、大量无用的 bloatware——修补不如重建。这是 Mio 的起点：一个为深度记忆和活人感而生的AI伴侣框架。