当用户即将基于二手信息(报告、传闻、他人的总结)做重大决策,或团队出现了"坐在办公室拍脑袋"的倾向时激活. An agent skill from kangarooking/mao-selected-works-skill.

MITAuto-check passed

Install Diaocha Yanjiu

skills CLI
$ npx skills add kangarooking/mao-selected-works-skill --skill diaocha-yanjiu -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install kangarooking/mao-selected-works-skill diaocha-yanjiu --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/kangarooking/mao-selected-works-skill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/diaocha-yanjiu .claude/skills/diaocha-yanjiu && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
diaocha-yanjiu
GitHub stars
115
Token cost
~905 tokens
SKILL.md length
109 words
Files
2
Skills in repo
13
Repo updated
First seen
Licence
MIT

At a glance

当用户即将基于二手信息(报告、传闻、他人的总结)做重大决策,或团队出现了"坐在办公室拍脑袋"的倾向时激活. An agent skill from kangarooking/mao-selected-works-skill.

  • Works in 6 steps: "没有调查就没有发言权"不是建议而是禁令——没有调查过的事,不配下结论。 → 调查的核心原则是"亲自做":(1) 亲自去现场,不是派人去汇报;(2)… → 调查要既听正面也听反面——只听一面之词比不调查更危险。 → …
  • SKILL.md covers R — 原文 (Reading), I — 方法论骨架 (Interpretation), A1 — 书中应用 (Past Application) and A2 — 触发场景 (Future Trigger), plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Diaocha Yanjiu is an agent skill from kangarooking/mao-selected-works-skill. 当用户即将基于二手信息(报告、传闻、他人的总结)做重大决策,或团队出现了"坐在办公室拍脑袋"的倾向时激活。 典型触发信号:"我看过几份行业报告了""派人去调研一下就行""凭我的经验判断应该是这样"。 不调用场景:决策已经基于充分的一手信息、低风险小决策不需要深度调研、已有可靠数据源在持续监测。 与"实事求是信息加工法"的区别:本skill关注"怎么采集高质量的一手信息",后者关注"拿到信息后怎么加工提炼"。

Its SKILL.md is about 910 tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `test-prompts.json`).

The repository describes itself as: An AI skill pack distilling Selected Works of Mao Zedong into reusable cognition, strategy, organization, and execution modules. The licence is MIT.

Example prompts

  • “坐在办公室拍脑袋”
  • “我看过几份行业报告了”
  • “派人去调研一下就行”
  • “/diaocha-yanjiu”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. "没有调查就没有发言权"不是建议而是禁令——没有调查过的事,不配下结论。
  2. 调查的核心原则是"亲自做":(1) 亲自去现场,不是派人去汇报;(2) 自己做记录,不假手于人;(3) 开调查会,找各阶层/各角色的代表,不只是成功者或友好对象。
  3. 调查要既听正面也听反面——只听一面之词比不调查更危险。
  4. 调查不是目的而是"入门"——"进门后还要抓本质"。调查获得感性材料后,还需"去粗取精"的加工功夫。
  5. 二手报告不等于调查——看报告是"吃别人嚼过的馍",信息已失真。创始人的手感来自一线。
  6. 调查是持续行为而非一次性动作——环境在变,认知需要持续更新。

What it can do on your machine

Read from SKILL.md and the folder at commit 5058fe4. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Diaocha Yanjiu loads about 905 tokens when it runs. Until then it costs about 55 tokens; SKILL.md has 109 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~55
When it runs · the whole SKILL.md, loaded when a task matches
~905

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from kangarooking/mao-selected-works-skill at commit 5058fe4, republished under its MIT licence (© kangarooking). 109 words, ~905 tokens.

Download SKILL.mdSave it as .claude/skills/diaocha-yanjiu/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
diaocha-yanjiu
description
当用户即将基于二手信息(报告、传闻、他人的总结)做重大决策,或团队出现了"坐在办公室拍脑袋"的倾向时激活。 典型触发信号:"我看过几份行业报告了""派人去调研一下就行""凭我的经验判断应该是这样"。 不调用场景:决策已经基于充分的一手信息、低风险小决策不需要深度调研、已有可靠数据源在持续监测。 与"实事求是信息加工法"的区别:本skill关注"怎么采集高质量的一手信息",后者关注"拿到信息后怎么加工提炼"。
source_book
《毛泽东选集第1-5卷》 毛泽东
source_chapter
反对本本主义(1930); 湖南农民运动考察报告(1927); 农村调查序言(1941)
tags
调查研究, 一手信息, 实地访谈, 反教条, 决策基础

R — 原文 (Reading)

"'没有调查就没有发言权',这句话,虽然曾经被人讥为'狭隘经验论'的,我却至今不悔;不但不悔,我仍然坚持没有调查是不可能有发言权的。有许多人,'下车伊始',就哇喇哇喇地发议论,提意见,这也批评,那也指责,其实这种人十个有十个要失败。" — 《农村调查》序言(1941年)

I — 方法论骨架 (Interpretation)

  1. "没有调查就没有发言权"不是建议而是禁令——没有调查过的事,不配下结论。
  2. 调查的核心原则是"亲自做":(1) 亲自去现场,不是派人去汇报;(2) 自己做记录,不假手于人;(3) 开调查会,找各阶层/各角色的代表,不只是成功者或友好对象。
  3. 调查要既听正面也听反面——只听一面之词比不调查更危险。
  4. 调查不是目的而是"入门"——"进门后还要抓本质"。调查获得感性材料后,还需"去粗取精"的加工功夫。
  5. 二手报告不等于调查——看报告是"吃别人嚼过的馍",信息已失真。创始人的手感来自一线。
  6. 调查是持续行为而非一次性动作——环境在变,认知需要持续更新。

A1 — 书中应用 (Past Application)

应用1:32天湖南实地调查逆转认知
  • 问题:1927年党内外对农民运动评价截然对立——绅士阶级说"糟得很",党内陈独秀倾向于压制农民运动。毛泽东在汉口、长沙从绅士阶级那里听到的道理与农民的实际情况完全相反。
  • 方法使用:毛泽东用32天实地调查湘潭、湘乡、衡山、醴陵、长沙五县,召集有经验的农民和农运工作同志开调查会,仔细听他们的报告,亲自做记录。
  • 结论:农民运动"好得很",不是"糟得很"。绅士阶级的判断代表了地主利益,不是真相。
  • 结果:调查结论被陈独秀拒绝,但历史证明毛泽东正确。这次调查成为"没有调查就没有发言权"的源头案例。
应用2:寻乌调查 — 深入一个点的解剖麻雀
  • 问题:需要了解中国农村经济的社会结构和阶级状况,但党内缺乏可靠的一手数据。
  • 方法使用:毛泽东做了著名的寻乌调查——深入一个县城,从商业状况到土地分配,从各阶级状况到日常物价,逐一详细调查记录。
  • 结论:通过"解剖一只麻雀"(深入一个典型),获得了对整个中国农村社会结构的认知框架。
  • 结果:寻乌调查成为中共调查研究的典范,其方法论影响了此后数十年的调查传统。

A2 — 触发场景 (Future Trigger)

  1. 即将基于二手报告做重大决策:CEO看了几份行业报告就准备进入新市场、推出新产品或做重大投资。用户会说:"行业报告显示这个市场有XX亿""分析师说趋势是这样的"。
  2. 团队坐在办公室拍脑袋:产品/战略决策完全基于内部讨论和数据分析,没有人去过一线。用户会说:"根据我们的数据分析""团队讨论的结果是"。
  3. 认知与市场反馈脱节:公司的判断与市场实际表现不一致,但没人去一线验证。用户会说:"用户应该会喜欢这个功能的""市场反馈和我们预想的不一样"。
  4. 空降高管/新进入行业:新来的人或新进入的行业,缺乏一手体感。用户会说:"我刚接手这个业务""这是我们第一次做这个领域"。
  5. 重大分歧无法裁决:团队内部对某个关键判断有重大分歧,各有各的数据和逻辑,但都缺乏一手验证。用户会说:"A团队说这样,B团队说那样,不知道谁对"。

语言信号:用户提到"报告""分析""别人说""数据显示""我觉得应该是""我们内部讨论"等词汇且没有提及一手调研时,应激活此skill提醒调查的必要性。

与相邻skill区分:

  • vs 实事求是信息加工法:本skill = 采集信息(拿到好的原料),后者 = 加工信息(把原料变成产品)
  • vs 群众路线:本skill是调查者个人的信息采集方法论,群众路线是组织层面的"从群众中来到群众中去"循环
  • vs 实践认识论:本skill是实践认识论中"获取感性认识"这一环节的具体操作方法

E — 可执行步骤 (Execution)

步骤1:明确调查问题和范围
  • 写下"我需要回答的核心问题是什么",以及"要回答这个问题我需要了解谁的看法/什么现状"。
  • 确定调查对象:不仅要有友好对象,还必须有反面意见持有者、一线执行者、终端用户。
  • 完成标准:产出调查清单:(1) 核心问题 (2) 需要调查的3-5类对象 (3) 每类对象至少2-3个具体人选。
步骤2:亲自到场、当面听取
  • 去现场,不是叫人来办公室。见终端用户,不是只听销售转述。做自己的笔记,不是让助理整理。
  • 开调查会时:先听少说,让对方充分表达;追问具体案例而非观点("能举个例子吗"比"你觉得怎么样"有效);既听正面也听反面("你觉得哪里不好"比"你觉得哪里好"信息量更大)。
  • 判停点:如果发现调查对象的回答高度一致且都是你想听的,可能样本偏差——必须补充反面对象和一线人员。
  • 完成标准:至少完成5个不同角色的深度访谈,记录中出现至少3个"与预想不同"的发现。
步骤3:从调查到判断
  • 将调查所得材料按"去粗取精"四步加工,提炼出对核心问题的回答。
  • 对比调查前后的判断变化:哪些假设被验证?哪些被推翻?哪些是新发现的?
  • 完成标准:产出"调查前判断 vs 调查后判断"对照表,且至少有1条重大认知修正。如果调查前后判断完全一致,说明调查深度不够。

B — 边界 (Boundary)

不要使用的场景
  1. 低风险小决策:不是每个决策都需要实地调查。日常运营的小决策用现有数据和经验即可。
  2. 时间极度紧迫:如果明天必须决策,来不及做完整调查。这时用已有的最好信息做判断,事后补调查。
  3. 已有高质量的一手信息流:如果公司已经建立了持续的一手信息采集机制(如高频用户访谈、一线轮岗制度),不需要每次都从头调查。
失败模式(来自反例)
  • 教条主义(ce01):调查后不加分析地照搬调查对象的意见——"用户说要X,我们就做X"。调查只是入门向导,进门后还要自己判断。
  • "钦差大臣"式调查(ce01变体):下去调查但带着预设结论,只搜集支持自己观点的证据,忽略反面信息。比不调查更危险。
  • 党八股式汇报(ce08):调查了很多但汇报时空话连篇、现象罗列——有调查的形式但没有调查的实质。"甲乙丙丁开中药铺"式罗列不等于分析。
作者盲点
  • 毛泽东晚年逐渐脱离一线调查,决策更多基于身边人的汇报而非亲自调查。大跃进时期相信了明显虚假的亩产数据,因为他没有亲自到农村去看。教训:即使是强烈主张调查的人,一旦停止亲自调查,判断力也会迅速下降。调查是持续行为,不能一劳永逸。

相关 skills

  • 群众路线 (composes-with): 调查研究是单次深度信息采集,群众路线是持续的"从群众中来到群众中去"循环。调查研究是群众路线的基础技能——先学会高质量的调查研究,再建立持续的群众路线循环。
  • 实事求是信息加工法 (composes-with): 调查研究法负责"采集高质量的一手信息"(原料获取),实事求是信息加工法负责"把原料加工成决策依据"(原料加工)。先调查采集,再实事求是加工。

© kangarooking, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in diaocha-yanjiu of kangarooking/mao-selected-works-skill.

  • SKILL.md
  • test-prompts.json

Open the folder on GitHubat commit 5058fe4

More from kangarooking/mao-selected-works-skill

All 13 skills in this repo
  • Buduicheng Zhanlue

    kangarooking/mao-selected-works-skill

    当用户作为弱势方面对强势竞争者、感觉"处处被动无法招架"时激活此skill. An agent skill from kangarooking/mao-selected-works-skill.

    115 GitHub stars~892 tokensUpdated 5 mo ago
    Auto-check passed
  • Chijiuzhan San Jieduan

    kangarooking/mao-selected-works-skill

    当用户面对明显强于自己的对手(巨头入场、资源差距、品牌劣势)需要规划长期竞争策略时激活. An agent skill from kangarooking/mao-selected-works-skill.

    115 GitHub stars~1k tokensUpdated 5 mo ago
    Auto-check passed
  • Jianmiezhan Jizhong Bingli

    kangarooking/mao-selected-works-skill

    当用户资源有限但面临多个机会/威胁,需要在多个方向间做取舍时激活。典型触发信号: "5个客户机会同时出现""要不要同时做ToB和ToC""资源不够分怎么办"。

    115 GitHub stars~1.1k tokensUpdated 5 mo ago
    Auto-check passed
  • Maodun Fenxi

    kangarooking/mao-selected-works-skill

    当用户面对多个相互交织的问题、无法判断优先级、陷入"什么都重要"的决策瘫痪时激活此skill. An agent skill from kangarooking/mao-selected-works-skill.

    115 GitHub stars~914 tokensUpdated 5 mo ago
    Auto-check passed
  • Maodun Techuxing

    kangarooking/mao-selected-works-skill

    当用户准备套用一个成功方法论(别人的或自己过去的)到新场景时激活此skill。典型触发信号: "某公司用了X方法成功了,我们也该试试""我之前这么做有效,继续做""行业最佳实践是Y"。

    115 GitHub stars~811 tokensUpdated 5 mo ago
    Auto-check passed
  • Shijian Renshilun

    kangarooking/mao-selected-works-skill

    当用户陷入"分析瘫痪"(想做但不敢行动)或"盲目行动"(做了但不总结不提升)时激活此skill. An agent skill from kangarooking/mao-selected-works-skill.

    115 GitHub stars~961 tokensUpdated 5 mo ago
    Auto-check passed

Questions about Diaocha Yanjiu

What does Diaocha Yanjiu do?

当用户即将基于二手信息(报告、传闻、他人的总结)做重大决策,或团队出现了"坐在办公室拍脑袋"的倾向时激活. An agent skill from kangarooking/mao-selected-works-skill. Diaocha Yanjiu is an agent skill from kangarooking/mao-selected-works-skill.

How do I install Diaocha Yanjiu in Claude Code?

Run `npx skills add kangarooking/mao-selected-works-skill --skill diaocha-yanjiu -a claude-code`. Or copy the skill folder (diaocha-yanjiu in kangarooking/mao-selected-works-skill) into .claude/skills/diaocha-yanjiu in your project. Claude Code loads it when a task matches its description.

How do I install Diaocha Yanjiu in Codex?

Run `npx skills add kangarooking/mao-selected-works-skill --skill diaocha-yanjiu -a codex`. Or copy the skill folder (diaocha-yanjiu in kangarooking/mao-selected-works-skill) into .agents/skills/diaocha-yanjiu in your project. Codex loads it when a task matches its description.

Can I use Diaocha Yanjiu in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add kangarooking/mao-selected-works-skill --skill diaocha-yanjiu -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/diaocha-yanjiu, .gemini/skills/diaocha-yanjiu, .github/skills/diaocha-yanjiu and .opencode/skills/diaocha-yanjiu in your project.

What does Diaocha Yanjiu need to run?

SKILL.md names no scripts, command-line tools or credentials: Diaocha Yanjiu is instructions for the agent only.

Does Diaocha Yanjiu access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Diaocha Yanjiu safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Diaocha Yanjiu use?

Diaocha Yanjiu is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Diaocha Yanjiu use?

About 905 tokens (SKILL.md is roughly 3.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

Who maintains Diaocha Yanjiu?

kangarooking (a GitHub user) maintains it in kangarooking/mao-selected-works-skill, which has 115 GitHub stars. The repository holds 13 skills in this directory. The repository was last updated on May 2, 2026.

Source: kangarooking/mao-selected-works-skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.