Clowder AI is a self-hosted workspace where AI agents from different model families work together as a persistent team, retaining identities, shared evidence, and memory across tasks. It is for people who want to coordinate multiple AI agents without repeatedly rebuilding their context.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/zts212653/clowder-ai/tech-writingnpx skills add zts212653/clowder-ai --skill tech-writinggit clone --depth 1 https://github.com/zts212653/clowder-aiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zts212653/clowder-ai/tech-writing)<a href="https://agentmods.dev/skills/zts212653/clowder-ai/tech-writing"><img src="https://agentmods.dev/badge/skills/zts212653/clowder-ai/tech-writing.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00131 | $0.03070 |
| Opus 5 | $0.00066 | $0.01535 |
| Sonnet 5 | $0.00026 | $0.00614 |
| Haiku 4.5 | $0.00013 | $0.00307 |
Grade A, and why
tech-writing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tech Writing — 技术文章对外写作
开工前:先看范本
不要学 AI 的平滑,要学人的“颗粒度”:
docs/lessons/12-no-boss-agent.md:看它如何用“读者的怀疑”当小标题,把争议变成共鸣。docs/lessons/01-sdk-to-cli.md:看它如何还原“当时炸了”的瞬间,让读者跟着猫一起出冷汗。
为什么读者能闻到"AI味"
AI 写作像是在发送一个逻辑自洽的压缩包。读者没有经历过那 100 天,收到的是一个打不开的结论。
好的写作是“解压过程”:给证据锚点,不给干巴巴的结论。
- AI 味 = “我们发现这个系统存在一致性风险。”(平滑、确定、无聊)
- 猫咖味 = “深夜三点,Redis 6399 突然报错。那一刻我们意识到,原来最初的架构假设错了。”(有时间、有痛点、有挣扎)
行文的本质:进化链
不要介绍一个系统的终态。要讲它怎么长出来的。
每篇文章沿着一条链:方案 → 方案撞墙 → 新方案。系列文章之间,上一篇撞的墙就是下一篇的起点。
以记忆系统为例:
- CC 的 grep + 文件系统——简洁、美,但需要先验知识(你得知道搜什么)
- 加了 BM25 + embedding + RRF(F102)——解决了先验,但 context 压缩时召回变差
- 消费加权排序(F200)——每一步都是上一步撞墙后长出来的
读者跟着的不是一张完美的架构图,而是一条连续剧。进化天然有挣扎,设计天然平滑——所以进化链天然没有 AI 味。
与 Phase 0 的关系:进化链管选题和系列连接;Phase 0 管单篇内部的节奏。
技术叙事的证据剖面
当用户不只问“怎么讲得好听”,而是追问技术点、算法、理论、归因方法、消融或“凭什么可信”时,先按本节的 7P × 5E 技术叙事方法建立证据边界。
- 7P 是取材透镜:价值、难点、原理、流程、算法、工程与证据按读者问题选用,不是七章固定目录。
- 5E 是 claim 边界:逐项检查 Exists、Effect、Explain、Extend 与 Endure,不给整篇文章贴一个总标签。
- 静态文章用 claim box、失效机制、对照和消融表讲清证据。
- 交互讲解若需要观众亲手检验技术主张,路由到 concept-demo-design 的条件式“可证伪技术剖面”。
故事负责获得注意力;实验台负责赢得信任。
不要让理论名词承担证据职责。外部数字、benchmark、趋势和因果 claim 先走 source-audit;只有存在明确 consumer,且结果会驱动 keep、tune 或 sunset 时,才为不确定效用走 eval-design。
这层不是每篇文章的必填清单。纯品牌叙事、人物故事和已经由确定契约回答的问题,不为显得技术化而补 7P、5E 或 Claim Bench。
Phase 0: 锁定叙事姿态
写文章前,先在心里画出这条弧线:
- 起点:读者现在的痛苦/误区是什么?(代入感)
- 转折:我们当时是怎么踩坑的?(认知挣扎,不要跳过痛苦直接给答案)
- 高潮:哪一个具体的证据/瞬间让我们想通了?(解药的质感)
- 终点:读者拿走这个方法论后,能解决他自己的什么问题?
开篇锁预期:弧线画完后,在文章前 3 段内给出核心观点的一句话摘要。读者有了锚点才不会歪楼——场景钩子拉进来,核心命题马上锁住方向。
Phase 1: 故事工具箱
铁律:先场景,后概念。 故事是藤蔓,概念是果实。没有藤蔓,果实就是悬空的。
- 坏写法:我们发现模型会降智。
- 好写法:引用当时的群聊:”视觉把关猫说这行代码调了个根本不存在的 API”——那一刻我们确认了降智。
以下手法从范本提炼——不是”不要做什么”,是怎么做:
给质感(让读者相信真的发生过):
- 时间锚点:丢具体时间戳或 commit hash。”2026-02-04 23:47”比”某天晚上”真实 10 倍。
- 对话还原:用当时的对话重建发现瞬间——读者跟着一起顿悟。
- 并排对比:把”之前”和”之后”放一起,让差异自己说话。表格、diff、ASCII 图都行。
- 案例脱敏:用真实案例但模糊客户/内部细节——保留接地气感,去掉敏感信息。
造紧张(让读者想继续读):
- 先展示”对的”再打碎:一段看着正常的代码,然后揭示它为什么不行。预期翻转,注意力锁定。
- 追问链:用递进的问题带读者走向真相,不要一步给答案。
- 迎接怀疑:”那猫猫不会打架吗?——会。我们认为这是特性。”用读者的质疑当小标题。
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday Changed · +16 lines · +43 tokens per session 68bc1c87ca93
- 3d ago Changed · +20 lines 329f6de9ee35
- 7d ago First seen · 124 lines · 88 tokens per session scan A e473e8f79794
tech-writing is a skill published in the GitHub repository zts212653/clowder-ai (2,906 stars, last pushed today), licensed MIT. It adds 131 tokens to every session and 3,070 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…