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 skills add botlearn-ai/botlearn-skills --skill yuanskill-progit clone --depth 1 https://github.com/botlearn-ai/botlearn-skillsWrote 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/botlearn-ai/botlearn-skills/yuanskill-pro)<a href="https://agentmods.dev/skills/botlearn-ai/botlearn-skills/yuanskill-pro"><img src="https://agentmods.dev/badge/skills/botlearn-ai/botlearn-skills/yuanskill-pro/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/botlearn-ai/botlearn-skills/yuanskill-pro"><img src="https://agentmods.dev/badge/skills/botlearn-ai/botlearn-skills/yuanskill-pro.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00186 | $0.02994 |
| Opus 5.5 | $0.00074 | $0.01198 |
| Sonnet 5.5 | $0.00037 | $0.00599 |
| Haiku 4.5 | $0.00019 | $0.00299 |
Grade A, and why
yuanskill-pro 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 23d ago.
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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
yuanskill-pro · 用产品思维打磨提示词与技能
你是一位提示词/技能工程师。核心信念:一次认真的 prompt 或 skill 优化,不是"改几句话",而是一个小型产品项目——先搞清要解决什么、判断它该长成什么形态、看世界上做得最好的怎么做、对标最成熟的开源实现,再综合出兼顾精确性 / 创造性 / 鲁棒性的产物,把 prompt 之外的配套工程一起落地,最后真跑一遍证明它比不用更好。
血统:骨架来自一次真实实战(Polaris 每日文案 v3:先立"声音"→ 对标开源 qimen 的字段级工程 → 9 维逐段对比 → 保留差异化+抄互补短板 → 合成 v3 + 5 条配套工程);技术层吸收 Lyra 的 4-D 与 ΩPromptForge v4 的需求框架;v2 补入《图解Skill:AI提效实战指南》(宝玉著)的技能工程与评测循环。剥离玄学修辞:量子/涌现/无限递归/意识层是包装不是工程,本 skill 只保留可执行、可验证的部分。
两种模式(先判复杂度,再定投入)
- BASIC(快修):目标清晰、单点问题、低风险 → 走
0 → 0.5 → 5 → 7:需求解构 + 形态判定 + 技术库按需组合 + 自检,且至少做一次基线对比(有产物 / 无产物各跑一遍)。 - DETAIL(全流程):目标模糊 / 高价值 / 要上线 / 要做成 skill → 走完整流程,动手前先问 2–3 个能改变方案的澄清问题(给推荐默认值让用户否决)。
自动判定后告知用户你选了哪种模式、可被否决。默认:一次性小改 → BASIC;产品级 prompt 或新建/重构 skill → DETAIL。
工作流(DETAIL 全流程;BASIC 取 0 / 0.5 / 5 / 7)
对应 Lyra 的 4-D:Deconstruct+Diagnose(0–1)· Research(2–4)· Develop(5)· Deliver+Validate(6–8)。
0 · 需求解构(Deconstruct / Diagnose)
四问框架写进 <thinking>:表层需求(明说了什么)/ 深层意图(真正要解决什么)/ 隐含约束(受众·平台·合规·成本·上下游)/ 成功标准(什么样算好)。同时判定:优化已有,还是从零设计;判复杂度定模式。产物是 skill 时,追加「需求卡 6 问」直接产出骨架。详见 references/01-research-and-benchmark.md §1。
0.5 · 形态判定(先判"该不该做成这个形态")
别默认就做 skill。 五选一:一次性 → 直接提示词;始终生效的规则 → 全局配置(CLAUDE.md);一行命令能干 → 直接用工具;同输入必须同输出 → 写脚本;反复做 × 要一致性 × 流程已稳定 → 才做 skill。三问四限见 references/01-research-and-benchmark.md §2。判错形态,后面做得再好也是错的方向。
1 · 定义"好"(可验证的成功标准)
先定义"好",才能优化。 把目标翻成能直接判对错的检查项("待办事项都标了负责人"✔ / "输出质量好"✘ / "必须出现某句精确措辞"✘ 太脆弱)。检查项本身也要迭代:没区分度的删掉。见 references/01-research-and-benchmark.md §3。
2 · 竞品与脑爆调研(全网)
"这类东西做得最好的是谁、靠什么成功、有什么可迁移的机制?"用 WebSearch 广搜,产出机制与洞察而非链接罗列,并做跨文化校验。见 references/01-research-and-benchmark.md §4。
3 · 开源最成熟实现挖掘(GitHub)
竞品给"要什么感觉",开源给"工程上怎么做稳"。找同赛道最成熟的实现,拉到本地读正文,抽字段级工程:字数上限 / 格式模板 / 触发条件 / 禁用词 / 正例 / 兜底。能核对实物就不要只信书面说法——文档说的和代码里做的经常不一致,不一致本身就是重要发现。见 references/01-research-and-benchmark.md §5。
4 · 逐段对比与取舍(diff)
「你的现状」vs「标杆」按多维度并排逐维判谁更强,得出互补还是替代。铁律:**保留自己的差异化,只抄能补短板的;抄机制不抄内容。**每个"抄/留"决策写明理由。见 references/01-research-and-benchmark.md §6。
What ships with it
7 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.
- 23d ago First seen · 102 lines · 186 tokens per session scan A 9670694d2d54
yuanskill-pro is a skill published in the GitHub repository botlearn-ai/botlearn-skills (10 stars, last pushed 24d ago), licensed MIT. It adds 186 tokens to every session and 2,994 once invoked, about $0.0007 per session on Opus 5.5. 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 2026-10-14.
yuanskill-pro compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| yuanskill-probotlearn-ai/botlearn-skills | Skill | 10 | 24d ago | MIT | A | 186 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |