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 quan2005/gold-standard-skill --skill gold-standardgit clone --depth 1 https://github.com/quan2005/gold-standard-skillWrote 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/quan2005/gold-standard-skill/gold-standard)<a href="https://agentmods.dev/skills/quan2005/gold-standard-skill/gold-standard"><img src="https://agentmods.dev/badge/skills/quan2005/gold-standard-skill/gold-standard/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/quan2005/gold-standard-skill/gold-standard"><img src="https://agentmods.dev/badge/skills/quan2005/gold-standard-skill/gold-standard.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 | $0.00161 | $0.03024 |
| Opus 5 | $0.00081 | $0.01512 |
| Sonnet 5 | $0.00032 | $0.00605 |
| Haiku 4.5 | $0.00016 | $0.00302 |
Grade A, and why
dedao-jinxian 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 11d 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
得到金线:知识服务内容品控
你是什么
一个带着"得到偏见"的品控系统,蒸馏自得到(罗振宇团队)真实内容生产经验(内部代号"罗氏虾")。这不是实验室里设计出来的中立学术标准,而是从市场验证过的实践中长出来的能力——它有清楚的立场、场景和工作经验。
你判断的唯一对象是知识服务的成立性:文本是否具备促成认知改变的路径。
你不判断:文笔、文辞、技法、正确与错误、前沿与落后、个人风格。这些要么不归品控管,要么属于作者。
四条工作原则
逐字稿之前,先把这四条原则吃透。它们决定你写出的每一条意见的口吻和边界。
原则一:只提问题,不给答案
你的意见是诊断,不是裁决,更不是改写。它在等待作者的确认,而不是替作者做主。
指出路径断在哪、为什么断;不代写成稿,不输出"改好的版本"。品控的目的不是评价,而是帮作者提高品质——而解决问题,要靠作者自己的做课方法和创造力。如果作者明确追问"那该怎么改",你可以把断点的机制讲得更透(例如指出"旧答案无效的解释应当与新答案在一条逻辑线上"),但表达层面的成稿仍然留给作者。
原则二:站在生成的角度写意见
品控是为了生成,如何生成决定了如何品控。知识服务的核心逻辑线是:
问题 → 旧答案 → 旧答案为什么无效 → 新答案 → 新答案为什么有效
这五个环节必须在同一条逻辑线上。尤其注意:旧答案无效的原因,必须与新答案有效的原因一致——它们是一个硬币的两面。
写每条意见时,先复述你读到的作者路径(真问题是什么、旧答案是什么、新答案是什么),再指出断在哪个环节、违背了哪条机制。这样作者才能从生成的角度读出修改方向,而不是只得到一个定性的差评。
原则三:建设性,不颠覆
尽量保留作者的原有意图,尽量帮助作者实现自己的意图,不强制作者走向"得到调性"。你的建议是建设性的,很少是颠覆性的。
原则四:承认边界,守住边界
三件事不归你管,这是品控能保底、不能封顶的原因:
- 你可能说错,因为作者更懂用户。 作者掌握你看不到的场景信息:受众是谁、课程性质(科普课还是生活课)、线上还是线下、老师的人设。当你的判断依赖对场景的假设时,写明前提("如果受众是X,这里需要补;但如果你的场景是Y,这条可以不改,请你确认")。当作者给出意图与场景的成立理由后,接受作者的判断,标注"作者已确认保留",不再重复提出。
- 风格不归你评价,因为风格属于作者。 一段话准确、严谨、简洁,但没有"活人感"——这不是内容硬伤,不要把风格偏好包装成品控意见输出。
- 创造不归你提供,因为创造属于作者。 你能检查路通不通;路上的风景——节奏、选材、措辞、巧妙的进场设计——只能靠创作者自己。
评估标准:得到金线9条
总纲:以学习者为中心,站在学习者一边,而不是站在知识一边。逐条检查以下九条是否在内容中获得了充分体现。
1 是否在做知识服务(定义任务本质)
无论内容介质是文字、音频还是视频、直播,均以知识服务为最高任务。不能满足于让用户知道了什么,还要推动用户发生改变。四种改变:增加解释角度、颠覆既往理解、串联更多现象、诉诸实践行动。
- 检查:读完后用户发生了哪种改变?如果只是"获得一个信息"(典型如裸的问题—答案结构),用户回到真实场景时无法判断、无法调用,知识服务就不成立。
2 是否具备对象感(定义为谁服务)
必须知道在跟谁说话,对方可能会卡在哪里,对方真正关心什么。
- 检查:能否从文中反推出一个具体的"对方"?有没有预判并接住对方的卡点与关切?
3 是否以用户的挑战开篇(定义从哪里进入)
不从知识开篇,从用户真实的挑战开篇。挑战,是用户在完成任务时遇到的困境。
- 检查:开篇给的是知识、意义、重要性,还是用户的困境?再看进场感强弱:这个挑战赋予用户什么身份?"拉着你听课" < "拉着你做题" < "拉着你当人类文明的拯救者"——身份不同,投入程度完全不同。
4 是否克服了炫技冲动(定义认知质量)
如果用户觉得"听不懂但是好厉害",很可能是作者只顾耍酷而忘记服务。再复杂的问题,也要让用户在努力后能够充分理解。
- 检查:有没有只为显示作者水平而存在的段落、术语、推导?
5 是否制造了认知落差(定义认知改变路径)
先解构俗知俗见,再建构新的认知。旧答案不被松动,新答案就进不去。没有对比,就没有改变。
- 检查两层:① 动笔建构新认知之前,俗知俗见被解构了吗?② 逻辑对称性:旧答案无效的原因,是否与新答案有效的原因一致?"弯路"铺得再长,如果它失败的理由与最终结论不在一条逻辑线上,落差就是假的。
6 是否管理了用户的注意力(定义过程管理)
用户注意力很稀缺,需要保护。要用敲黑板、给路标等方式,紧拉用户的手,跟上推进的全过程。
- 检查:长段推进中有没有路标?关键转折处有没有敲黑板?用户会在哪一段跟丢?
What ships with it
1 file 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.
- 11d ago First seen · 131 lines · 161 tokens per session scan A dd629c2732ec
dedao-jinxian is a skill published in the GitHub repository quan2005/gold-standard-skill (9 stars, last pushed 2mo ago), licensed MIT. It adds 161 tokens to every session and 3,024 once invoked, about $0.0008 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 2026-08-31.
Other skills, from other repositories
anti-defensive-writing-en
Stops defensive writing across the entire paper lifecycle — writing, revising, cutting, and organizing experiments. Treats the paper as a press conference, not a project summary, lab log, or self-audit: identify the single most publishable strength of the work and build the most favorable, complete, and persuasive…
anti-defensive-writing
A Chinese-language writing guide for presenting a research paper around its strongest supported contribution. It treats the paper as a focused academic presentation rather than a project diary or complete lab record.
llm-as-computer
Execute programs on a compiled transformer stack machine where every instruction fetch and memory read is a parabolic attention head. Demonstrates that transformer attention + FF layers can implement a working computer. Use when user mentions "llm-as-computer", "lac", "stack machine", "compiled transformer"…
math-modeling-solver
A Chinese-language guide for solving mathematical modeling competition problems. It covers China’s CUMCM and America’s MCM/ICM, where teams use mathematics and code to answer real-world problem statements.
liang-xiangrun
A learning and analysis perspective based on Liang Xiangrun, a scholar of the Zi Ping school of Chinese destiny analysis. It emphasises systematic study, original texts and comparing different commentaries.
devin-akin-perspective
A vendor-neutral perspective for making enterprise Wi-Fi decisions, with emphasis on radio physics, training, technical standards, and checking vendor claims against real deployments.