Skip to content

xyaz1313/xyskill / skills/xy-ai-workflow

xy-ai-workflow

Verified

【AI流水线】内容生产 AI 流水线——先审计你现在从想法到发布每一步是谁在做、卡在哪,再判定哪一步该给 AI、哪一步必须留人,然后把想法→选题→逐字稿→切片→标题→发布排雷编排成一条有交接物的流水线,教你"某个环节 AI 跑通了就固化成 skill"的方法,并划清文风复制的正当边界(复制结构,不复制别人的稿)。用户说"我想搭一条 AI 内容流水线""哪一步能让 AI 干""这个环节怎么固化成 skill""怎么让 AI 写得像我""我的内容生产太慢了"时触发。|小爷出品

FileKindSize
SKILL.mdskill md35.8 KB
agents/openai.yamlconfig319 B
evals/evals.jsonconfig2.3 KB
references/atoms.jsonlother314.6 KB
scripts/atoms-search.pyexecutablescript11.5 KB

references/atoms.jsonl

View on GitHub
{"id": "XY-BKG-1556", "type": "method", "confidence": "medium", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "用一个小、低门槛、人人都能秒懂的日常消费决策(笔记本电脑选Air还是Pro)作为引子,先用具体的实操证据(自热板贴纸摸起来是凉的)解决这个小决策,提炼出一句好记的原则(不需要买贵的,一定要买对的),然后把这个原则直接嫁接、类比到真正想讲的抽象/专业话题(怎么选AI agent)上。这个结构的价值在于:抽象决策原则(如何选工具)本身很难讲得有代入感,但用一个几乎每个人都经历过的消费纠结(该买贵的还是够用的)先建立情绪认同,再做一次话题跳转,观众会更容易把刚认同的原则套用到陌生领域,理解成本和抵触感都更低。这条原始记录内容本身比较简短、偏个人经验分享,但这个「小消费决策类比+原则提炼+类比映射到抽象话题」的三段式结构是可复用的逐字稿节奏手法,值得单独提炼;内容本身不算强案例(缺少具体数据支撑,论证略单薄,接近软广),但结构本身有复用价值。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "逐字稿节奏", "blocks": ["12-4", "04-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 7}
{"id": "XY-KS-010", "type": "case", "confidence": "high", "topics": ["团队与模式设计", "认知与心态", "AI与工具"], "knowledge": "放手团队的真实案例:2020 年底判断团队成熟后选择放手、只身赴上海开新盘,原盘照跑(社区店后来发展到 300 多家)——放手的前提不是「我累了」而是「团队被验证过了」:疫情三个月线上 2000 万那一仗就是验证。放手后开的是新业务不是躺平。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["11-6", "13-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-mode", "xy-ops"], "related": null, "q": 6}
{"id": "XY-SY-0130", "type": "method", "confidence": "medium", "topics": ["私域运营", "流量获取", "AI与工具"], "knowledge": "AI视频素材矩阵打法:利用生成式 AI 视频工具(具体用哪家按当下可用的选,工具更替快、方法不变),通过克隆真人形象(Face Cloning)结合精准提示词(Prompt),快速批量产出高质量IP口播视频。核心逻辑在于:1. 形象克隆确保IP一致性;2. 提示词需包含“人物+场景+具体行为”三个维度;3. 文案控制在15秒内以适配短视频节奏。这种方法能极大地降低IP内容的生产成本,实现素材的规模化产出。在实际生意场景中,应将AI视频视为“消耗品”,其核心目标不是追求单条视频的艺术性,而是通过高频、多样的素材矩阵进行“种草”,最终导流至直播间或私域实现转化。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["12-4", "06-11"], "skills": ["xy-ai-workflow", "xy-ip", "xy-ops", "xy-playbook", "xy-private-ops", "xy-traffic"], "related": null, "q": 6}
{"id": "XY-DY-021", "type": "case", "confidence": "high", "topics": ["IP人设", "商业案例与实战复盘", "AI与工具"], "knowledge": "19岁少年用「退学+900元闯上海+一年赚到人生第一个100万」的人设定位案例——靠贩卖信息差(帮人安装海外AI工具,收费100-500元/单)赚钱,本质是抓住了「1%的人已经会用AI工具,99%的人不会用也不愿花时间学」这个巨大信息差", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["13-1", "07-1"], "skills": ["xy-ai-workflow", "xy-ip", "xy-ops", "xy-precedent"], "related": null, "q": 5}
{"id": "XY-DY-034", "type": "case", "confidence": "high", "topics": ["流量获取", "IP人设", "内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "19岁英国少年用AI零代码开发「30天戒色挑战」App案例——发现健身赛道几十款App但戒色这个高痛点需求赛道空白,配合个性化评估+游戏化通关+行为监测干扰三个设计,找健身网红做「戒色=变强者」的定位营销,3个月10万用户年收入300万美金", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["13-1", "07-6"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-ops", "xy-precedent"], "related": null, "q": 5}
{"id": "XY-DY-035", "type": "case", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "美国大学生用AI连续做3款单点需求App年入850万美金案例——从自己刷社群发现「直男不会高情商回复」的痛点做出回复生成器App一炮而红,再顺势推导出「穿搭建议」「卡路里识别」两款关联App,核心方法是从自己生活的真实困扰里找需求", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["13-1", "12-1"], "skills": ["xy-ai-workflow", "xy-ops", "xy-precedent", "xy-private-ops"], "related": null, "q": 5}
{"id": "XY-DY-277", "type": "case", "confidence": "high", "topics": ["选品逻辑", "商业案例与实战复盘", "AI与工具"], "knowledge": "产品营销闭环案例——某17岁少年开发的AI写作预测工具,靠「让产品自己秀肌肉」(发布A/B测试demo推文获330万浏览量,把自己打造成使用案例活招牌)+「24小时销售马拉松直播」+「效果不达标全额退款保证」三个动作,5小时卖出1万美元、24小时吸金3万美元;核心逻辑是先用真实效果证明可信度,再用高强度直播压缩决策周期,最后用退款保证消除购买顾虑", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["13-1", "05-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ops", "xy-precedent", "xy-selection"], "related": null, "q": 5}
{"id": "XY-WZ-079", "type": "case", "confidence": "high", "topics": ["私域运营", "团队与模式设计", "认知与心态", "内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "「AI青少年未来成长科创营」是一个3天时长的产品案例,内容设计分DAY1启智之旅(名企探访、院士领学)、DAY2锻造之旅(AI造物、AI竞赛、AI创业)、DAY3巅峰之旅(成果展示、创业赋能、圆满结业)三个递进阶段;权益设计上分别为孩子(科创思维/AI技能、赛事技能、创业思维、懂人才培养、科研履历)和父母(亲子共学、塑AI认知、懂人才培养)两类人群单独设计了不同的收获点,同时展示师资团队背景(如北大信研院背景)来做信任背书", "source_type": "account_video", "source_label": "小爷账号-课程", "chapter": null, "section": null, "blocks": ["07-6", "13-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-content-scan", "xy-mode", "xy-ops", "xy-precedent", "xy-private-ops"], "related": null, "q": 5}
{"id": "XY-KS-002", "type": "case", "confidence": "high", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "负债翻身案例:广告公司时期为垫资货款陷入三角债、信用卡套现周转,2012 年滚到 100 多万债务;孩子出生前托美国亲戚代购奶粉奶瓶,随手一条朋友圈意外接到 200 支奶瓶订单,抓住后 2012-2013 全力做美国代购,靠蚂蚁搬家式执行把 100 多万债务还清——转机不是想出来的,是在真实需求撞上门时接住的。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["01-2", "13-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-kickoff", "xy-precedent", "xy-private-ops"], "related": null, "q": 5}
{"id": "XY-KS-004", "type": "case", "confidence": "high", "topics": ["选品逻辑", "商业案例与实战复盘", "AI与工具"], "knowledge": "风口切换案例(化妆品→大健康):2015 年判断化妆品赛道严重饱和(全国一年新增 3000 多个品牌,红海杀不出未来),转身切入当时极少人关注、市场还排斥的中医养生足贴,从 2015 年 1 月起盘到年底 12 个月做了 1.25 亿营收,后用利润买地建厂实现自产自销——选未被教育的市场,吃的是先手教育红利。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["01-4", "13-1"], "skills": ["xy-ai-workflow", "xy-playbook", "xy-precedent", "xy-selection"], "related": null, "q": 5}
{"id": "XY-DY3-002", "type": "anti-pattern", "confidence": "high", "topics": ["私域运营", "合规与风控", "AI与工具"], "knowledge": "别用 OpenClaw 这类开源 agent 框架直接去接微信,市面上那些\"接上了\"的演示全是炫技。检验标准只有一个:能不能扛一年、两年不封号,没这个能力就别碰——号才是最重要的资产,封号比什么都痛。做私域的人自己验证过的稳妥路线:用扣子搭一个智能体接在微信里,真实扛了一年不封。想玩可以接小红书、飞书这类平台练手;而且企微很快会有官方的 agent 接入,微信也将内测自家方案,等官方端口比冒险用第三方 hook 划算。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "你们想多了", "blocks": ["12-4", "08-8"], "skills": ["xy-ai-workflow", "xy-private-ops", "xy-publish-guard"], "related": null, "q": 5}
{"id": "XY-DY3-013", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "蹭大模型限免活动(如爱马仕×小米 MIMO 的两周免费)有四个实测出来的坑:①免费额度走活动专属端口,不能用模型厂商官方平台原有的 api key——实测从小米 mimo 官方 key 接入,跑了一天全在扣费,问了 AI 才发现要换端口;②活动端口拿 key 前通常要先做最小订阅(0.01 那档)再充 10 刀门槛,不然拿不到 key;③配置一定调 Claude Code 来做,别让 agent 自己配、自己换模型,它会把自己换死;④给 agent 设定\"10 刀免费额度烧完自动切回付费 key\",免费期结束无缝过渡。模型搭配:主模型用 V2 Pro,图片识别单独走 V2 的 Omni 模型,语音 TTS 日常用不到。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "MIMO2周限免活动你们知道吗", "blocks": ["04-3", "12-3"], "skills": ["xy-ai-workflow"], "related": null, "q": 5}
{"id": "XY-BKG-1560", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "在讲一个技术性很强的话题(架构设计)之前,主动、具体地承认自己的能力短板(不是工程师、不是码农、代码看不懂),而不是回避或模糊带过,然后用一句朴素的话补上自己的优势(发挥想象力,其他交给AI)。这种「先自曝短板、再给出朴素的替代能力」的表达方式,是真人说话区别于AI生成腔调的典型特征之一——AI生成的营销/教程文案通常会强化权威感、很少主动承认具体的专业缺陷;而真实的人在分享经验时会自然带出「我其实不懂这个」这类具体的自我限定,反而增强可信度和亲近感。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["06-7"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 5}
{"id": "XY-KB-060", "type": "method", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "传统的「账号+重量级内容+收割」式引流模型仍有效但正在低效化,原因有三:用户心智水平提高难被简单内容击穿、同质化严重稀释效果、平台进入闭环存量阶段规则趋严。升级方向是操作流程去人工化:一是用AI工具压内容成本,一个精通AI训练的编导人效可提升5倍以上;二是引入群控SAAS系统和定制化CMS系统替代人工手撸操作,操作效率提高3-5倍;三是重建更高效的内容生产流程。前沿团队已组建专门的AI效率部门,把AI系统性嵌入各环节。这种升级不只是省钱,而是把整个运营体系拔高一个维度。", "source_type": "user_import", "source_label": "小爷3.0克隆知识库-202608", "chapter": null, "section": "创业者入局思考", "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-traffic"], "related": null, "q": 5}
{"id": "XY-KS-012", "type": "case", "confidence": "medium", "topics": ["私域运营", "成交与话术", "AI与工具"], "knowledge": "低价课的定价逻辑:线上课定价 399 元,明确不是引流款套路,是知道目标人群(创业兄弟)手头紧的照顾式定价;后端接线下课(12 期,聚会式:喝茶聊天对接资源)——低价可以成立,前提是它是真实交付而非转高价的钩子,且后端靠关系与资源链接而非话术逼单。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["09-2", "08-7"], "skills": ["xy-ai-workflow", "xy-close", "xy-opener", "xy-private-ops"], "related": null, "q": 5}
{"id": "XY-DY3-012", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "给 agent 接飞书比接微信省心,用爱马仕接飞书比折腾龙虾简单得多(原话:比龙虾简单100倍)。飞书开放平台建应用时,卡点集中在四处:①事件回调配置里要加上\"接收消息\"权限并保存;②要先添加机器人能力,应用才收得到消息;③权限管理用批量导入一次开满(可直接导入177项权限),别一条条点;④配完必须\"创建版本并发布\",应用不发布就没法用,看到绿色的发布成功才算数。原则是先让最小链路跑通、能收发消息,连聊天都不通就谈不上别的,细节权限之后再回头补。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "爱马仕接飞书他来了", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 5}
{"id": "XY-DY3-029", "type": "number", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "Token 成本可以做成单独的解决方案卖点:专门开发省 Token 的 Skill,实测基本能帮用户省出接近 50%~60% 的 token 消耗,且不影响原本的功能使用——后者正是用户最担心的点。token 消耗是 Agent 重度使用者的第一痛点,把\"降本且不减功能\"做成一个可直接安装的 Skill 单独发布,是最快能交付的一类工具产品。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "一早同时开发3套AI系统", "blocks": ["12-3", "10-5"], "skills": ["xy-ai-workflow"], "related": null, "q": 5}
{"id": "XY-DY3-030", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "九成人裸用 Claude Code 所以总翻车——不是模型不行,是让一个 AI 同时读项目、找 bug、写代码、做审查、补测试,它当然容易崩。正确用法是任务分工,装六个 Skill 各司其职:Explore 先画项目地图(入口在哪、核心文件在哪、功能调用链);Debug 拿到日志、复现步骤和最近改动后先定位根因,再给最小修复方案而不是乱修;Code Review 在改完一大段后挑逻辑漏洞、边界条件和回归风险;Test Engineer 判断哪里最容易回归、哪些测试最值得补;Code Simplifier 删重复逻辑、收命名收结构但不改变行为;Security Review 在涉及登录权限、密钥、支付、用户数据时最后扫一遍。高手不是让 AI 从头干到尾,而是把不同任务交给不同 Skill。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "Claude Code新手必装6个Skill", "blocks": ["12-4", "06-3"], "skills": ["xy-ai-workflow", "xy-ip", "xy-workbench"], "related": null, "q": 5}
{"id": "XY-SY-0052", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "私域防封风控操作SOP:通过模拟真人行为来降低平台AI监管风险。核心动作包括:1. 避免SOP式的话术机械复制;2. 控制加人节奏,主动添加或被动通过均应分时间间隔进行(如单次通过控制在5-10人),避免短时间内大规模操作;3. 强化互动闭环,加人后的首句沟通必须通过“钩子”诱导对方回复(如语音条或提问),触发双向互动以建立账号权重;4. 严格控制群发频率,建议将单次大额群发拆分为小规模、长间隔操作(如每次30-50人,间隔2-3小时)。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["08-9", "03-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-opener", "xy-ops", "xy-private-ops"], "related": null, "q": 5}
{"id": "XY-SY2-0507", "type": "method", "confidence": "medium", "topics": ["IP人设", "AI与工具"], "knowledge": "“1人公司”轻资产创业法:指利用AI工具、信息差及个人IP,将个体能力产品化,实现从零到一的商业闭环。该方法的核心在于降低组织成本,通过高度自动化的流程(如AI辅助内容生产)替代传统团队。在生意场景中,这意味着创业者不应过早追求规模化雇佣,而应优先构建“个人能力+数字化工具”的最小可行性业务模型(MVP),通过实时记录真实创业过程来建立信任,实现商业逻辑的快速验证。", "source_type": "course", "source_label": "小爷账号-2026私域星聚课(配图与课件)", "chapter": null, "section": "课程配图内容.md", "blocks": ["12-3", "07-5"], "skills": ["xy-ai-workflow", "xy-ip", "xy-mode", "xy-ops"], "related": null, "q": 5}
{"id": "XY-DY-009", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "评估AI工具/账号投入是否值得的标准:不是看工具本身贵不贵,是看它的产出能不能远超投入成本(例:账号成本约150-160元,能不能换来500-1000元的产出结果)", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["04-1"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 4}
{"id": "XY-DY-145", "type": "method", "confidence": "high", "topics": ["选品逻辑", "团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "团队成本要跟着项目ROI和利润动态扩张而不是一次性堆人——测试期先精简人员甚至用阿米巴机制、AI提效减少人力依赖,先让项目跑通盈利再扩张团队,避免一上来配五六人团队+设备+场租(一测试就是大几十万成本),项目没做成这些成本就全打水漂", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["11-6", "10-2"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-precedent", "xy-selection"], "related": null, "q": 4}
{"id": "XY-DY-195", "type": "case", "confidence": "high", "topics": ["流量获取", "内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "互联网内容原创/版权争议意义不大——粉丝只关心内容对自己有没有用,内容互相借鉴是行业常态,做自媒体的核心能力是「做流量的能力」而不是「做产品的能力」;案例:两个做同类AI智能体产品的人,擅长做流量+跟粉丝关系好的一方靠开课已赚五六百万,另一方产品能力更强但几乎没有流量和收入", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["03-2", "06-10"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ops", "xy-precedent"], "related": null, "q": 4}
{"id": "XY-DY-196", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "把复杂专业内容(如AI技术)翻译成普通人能听懂的语言,是获得大众认可并实现商业转化的关键技巧——纯专业角度讲清楚不等于让普通人听懂,能不能让老百姓听懂直接决定了转化上限", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["06-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ops"], "related": null, "q": 4}
{"id": "XY-DY-198", "type": "method", "confidence": "high", "topics": ["私域运营", "流量获取", "选品逻辑", "IP人设", "认知与心态", "内容创作与平台", "成交与话术"], "knowledge": "起号标准化SOP的核心是把选题/文案测试和执行拍摄分开——由专门小组批量测试各种选题文案找到能破10万播放的爆款,再交给60分IP直接复制拍摄(复制版本能到20万播放就算起号成功),60分IP本人不需要具备内容创作能力,只需要执行力和露脸配合,7天左右即可完成一个账号的标签打垂直", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["06-10", "03-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-content-scan", "xy-ip", "xy-ops", "xy-private-ops", "xy-script-glue", "xy-selection"], "related": null, "q": 4}
{"id": "XY-DY-238", "type": "method", "confidence": "high", "topics": ["私域运营", "团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "运营杠杆的两个关键抓手是AI提效和上SCRM系统——用AI+系统把原本需要10人的团队压缩到3人+AI就能搞定,私域圈90%的团队还没有上SCRM系统去精细化管理客户数据", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["12-4", "08-3"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-precedent", "xy-private-ops"], "related": null, "q": 4}
{"id": "XY-WZ-103", "type": "method", "confidence": "high", "topics": ["IP人设", "内容创作与平台", "AI与工具"], "knowledge": "高净值、强监管的赛道要以最低成本且不占用IP本人精力的方式达成近乎垄断的内容覆盖,具体做法是把画面与音频分离——用AI配音搭配任意可用的画面素材,突破了「必须由IP本人出镜/录音」的产能瓶颈,实现了内容生产的无上限复制", "source_type": "account_video", "source_label": "小爷账号-课程", "chapter": null, "section": null, "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-ops"], "related": null, "q": 4}
{"id": "XY-WZ-104", "type": "case", "confidence": "high", "topics": ["IP人设", "内容创作与平台", "AI与工具"], "knowledge": "把「搭建好工作流、运用好AI」落到具体产能上:单人运营,一天能有效生产100条以上内容——这个产能水平使得原本因为敏感而不敢进入的赛道变得可以尝试,也让原本难以规模化的IP有可能被放大成品牌矩阵", "source_type": "account_video", "source_label": "小爷账号-课程", "chapter": null, "section": null, "blocks": ["06-10", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-ops"], "related": null, "q": 4}
{"id": "XY-TW-013", "type": "case", "confidence": "high", "topics": ["AI与工具"], "knowledge": "充值699元买某AI编程订阅三天用完的真实经历:套餐里编程类agent享受高倍额度,但通用agent吃的是月度总池,没有高倍这回事;不知道这个区别、天天开着通用agent跑,三天就把编程额度也一起拖没了,且月度总池打空后要等下个月,中间没法用。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-19 12:12", "blocks": ["12-3", "08-8"], "skills": ["xy-ai-workflow"], "related": null, "q": 4}
{"id": "XY-TW-028", "type": "number", "confidence": "high", "topics": ["AI与工具"], "knowledge": "一个开源的Agent插件(路由分档/风险防线/三层记忆/自愈系统)交付前跑到8个模块、166个测试全过,全程走官方插件钩子接入、不改核心代码一行,作为\"开源前该做到什么程度\"的一个真实参照点。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-02 14:33", "blocks": ["12-4", "13-1"], "skills": ["xy-ai-workflow", "xy-opener"], "related": null, "q": 4}
{"id": "XY-TW-039", "type": "anti-pattern", "confidence": "high", "topics": ["AI与工具"], "knowledge": "\"Prompt已死,现在是循环工程/图工程时代\"这类说法是讲给1%人听的话——判断是否真的需要上循环工程,要同时满足四个硬条件:任务至少每周重复一次、有自动化验证手段(不是靠人眼看)、token预算烧得起、agent有真工具能自己跑代码看结果;四条不全中就不值得,绝大多数创作者和中小企业用不上这类重型工具,却被\"已死\"两个字架着追新。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-31 00:22", "blocks": ["12-4", "12-3"], "skills": ["xy-ai-workflow"], "related": null, "q": 4}
{"id": "XY-KS-020", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "内容工具的「Step 0 定位」法:使用者只填一次自己的定位(行业、人群、内容调性、差异点),之后所有内容生成都自动落进这个定位的场景,工具本身不带任何预设行业视角;没填就先问齐这四格再干活,不许跳过——防止产出「谁都能用等于谁都不贴」的通稿。", "source_type": "user_import", "source_label": "小爷早期Skill资产-20260820", "chapter": null, "section": "爆款机", "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-coach", "xy-content-scan", "xy-ip", "xy-opener", "xy-xhs-headline"], "related": null, "q": 4}
{"id": "XY-KS-003", "type": "anti-pattern", "confidence": "high", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "代购团队失控教训:跟着做代理的人超过 1000 人后,因代购无门槛,有人拿他朋友圈正品实拍图转卖高仿;同时海关物流 7 天到 3 个月不稳定导致退货——没有管理的团队就是一盘散沙,交付时效不稳会直接杀死高客单信任。后来靠转型(香港人肉代购保时效 + 投奔有管理体系的微商学招商团队裂变)解决。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["11-5", "13-2"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-precedent", "xy-private-ops"], "related": null, "q": 4}
{"id": "XY-KS-005", "type": "anti-pattern", "confidence": "high", "topics": ["流量获取", "认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "200 万买来的时机教训:2017 年预感短视频是风口,砸 200 万找了 5 家专业团队打造抖音 IP,全部打水漂——不是方向错了(方向后来被验证),是商业环境不成熟、自己也没想明白就重金入场。一定要在对的时间干对的事:太早入场和不入场一样亏,区别是太早还搭进本金。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["01-4", "13-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-ip", "xy-kickoff", "xy-peer-pick"], "related": null, "q": 4}
{"id": "XY-KS-006", "type": "anti-pattern", "confidence": "high", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "重资产模型不可复制教训:2017 年做 300 多平大健康直营店,盈利模型跑通了但单店装修近 30 万(不含店租人力),经销商根本无法复制——模型能赚钱和模型能复制是两回事,B 端生意里「经销商复制得起」比「单店利润好看」重要。改成社区小店磨了一年,2022 年后才在全国开出 300 多家社区加盟店。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["11-1", "13-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-mode", "xy-ops"], "related": null, "q": 4}
{"id": "XY-KS-007", "type": "case", "confidence": "high", "topics": ["私域运营", "商业案例与实战复盘", "成交与话术", "AI与工具"], "knowledge": "私域功底的极限测试:2020 年疫情封控、全年线下招商会泡汤、三个月完全发不出货,团队全部调回线上,在客户只能先交钱、货后补的条件下做了 2000 万业绩——私域的本质是信任储备,平时攒的信任在完全无法交付的极端条件下也能兑现成交。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["08-7", "13-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-precedent", "xy-private-ops"], "related": null, "q": 4}
{"id": "XY-KS-008", "type": "case", "confidence": "high", "topics": ["选品逻辑", "商业案例与实战复盘", "AI与工具"], "knowledge": "二次验证「未被教育市场」打法:2020 底赴上海创精油品牌,判断依据与足贴同构——芳疗是又一个未被完全教育的市场;普通人不懂配方不懂君臣佐使,痛点即切入点:直接做复方精油给用户最简单的解决方案,两年半做了 2 个多亿营业额。同一个选品逻辑在不同品类第二次成立。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["10-6", "13-1"], "skills": ["xy-ai-workflow", "xy-playbook", "xy-precedent", "xy-selection"], "related": null, "q": 4}
{"id": "XY-KS-009", "type": "case", "confidence": "high", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "轻资产小团队模型:回厦门做青梅精供应链,只做品牌代工与运营服务,公司仅 5 人服务五六个大客户,一年 3000 多万业绩、毛利约 20%——对 5 人团队是高人效高舒适度的生意。做重做轻是选择:经历过买地建厂后主动选轻,人效比规模优先。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["11-6"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-mode", "xy-ops", "xy-precedent"], "related": null, "q": 4}
{"id": "XY-KS-011", "type": "case", "confidence": "high", "topics": ["流量获取", "IP人设", "内容创作与平台", "AI与工具"], "knowledge": "IP 起号实录:2023 年 11 月因看不下去网上讲私域的误人子弟而开始写文章纠正错误概念,第一篇文章即 10 万+ 阅读;一条「一个月 200 万私域盘怎么做」的视频拿到 55 万自然流播放彻底破圈——凭十几年实操积累的「纠错型内容」切入,比追热点更能立住专家位。", "source_type": "user_import", "source_label": "小爷公众号自述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["07-5", "13-1"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-opener"], "related": null, "q": 4}
{"id": "XY-KS-014", "type": "case", "confidence": "high", "topics": ["商业案例与实战复盘", "成交与话术", "AI与工具"], "knowledge": "30 万学费回本实录:入场后花 30 万找 8 个老师学 AI(扣子、Prompt,Claude Sonnet 3.5 时代);2025 年 3 月决定把学费赚回来——3 月 15 日宣布开课,3 月 29 日就开出第一期,30 人 × 1 万元,30 万学费当期回本;此后每月一期 AI 线下课延续至今,客单价 1 万。学习投入要设计成产品闭环:学费不是沉没成本,是首期课程的进货成本。", "source_type": "user_import", "source_label": "小爷口述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["13-1", "12-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-kickoff", "xy-precedent"], "related": null, "q": 4}
{"id": "XY-KS-016", "type": "definition", "confidence": "high", "topics": ["IP人设", "商业案例与实战复盘", "AI与工具"], "knowledge": "小爷的品牌矩阵(均出自本人经历,可在内容与课程中使用):添行健(中医养生/足贴,2015 起盘 12 个月 1.25 亿)、安的种子(复方精油,上海两年半 2 亿+)、麦铛铛科技(大健康流量孵化器,2025 上海)、私域星聚(私域线下课,12 期)、风物执守(AI 线下课,2026-08-20 定名;取名思路同「生物股长」式——不解释品类,以一件具体风物立牌)。", "source_type": "user_import", "source_label": "小爷口述-20260820", "chapter": null, "section": "品牌矩阵", "blocks": ["07-3", "13-3"], "skills": ["xy-ai-workflow", "xy-coach", "xy-ip", "xy-term-crack"], "related": null, "q": 4}
{"id": "XY-DY3-006", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "装 OpenClaw(龙虾)时要同时装一个 Claude Code 给它做兜底,代码层面的事全让它干。原因:龙虾在自己体内升级很容易把自己搞挂,而 Claude Code 有底层代码能力,可以把龙虾的全部源码(1884 个文件)读完后边读边实施、一步步升级,双向打通后龙虾对升级过程还有体感、能反馈。分工原则就是\"专业的人干专业的事\"——除非你龙虾大脑接的模型本身接近 Claude 级别(代码与中文能力都顶级),否则把\"干活的 agent\"和\"管代码的 agent\"分开,是最稳的结构。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "用源码来升级openclaw龙虾", "blocks": ["12-4", "12-3"], "skills": ["xy-ai-workflow", "xy-workbench"], "related": ["XY-DY3-009"], "q": 4}
{"id": "XY-DY3-007", "type": "number", "confidence": "high", "topics": ["AI与工具"], "knowledge": "本地部署 agent 别买\"龙虾盒子\"这类 AI 盒子产品,纯割韭菜,直接买苹果电脑就够:预算紧就买二手 Mac,2020 年的老机器用了六年照样好用;经常移动办公就买新出的 M5 笔记本,约 1 万元;基本不关机、固定在办公室或家里用,就买 Mac mini,4000 多、国补后约 3800-3900,配上显示器键鼠总预算五六千搞定;M4 版 3000 多也不卡。选苹果的原因是架构:它的 CPU 和 GPU 共享统一内存,跟 Windows 完全不同,跑 AI 更合适;这类设备玩 AI 基本够用。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "玩龙虾别被乱割韭菜", "blocks": ["12-1"], "skills": ["xy-ai-workflow"], "related": null, "q": 4}
{"id": "XY-DY3-014", "type": "case", "confidence": "high", "topics": ["AI与工具"], "knowledge": "agent 全自动接活赚钱的真实编排:让爱马仕(Hermes)自己上 GitHub 找悬赏任务,看到简单的顺手写代码交 PR,任务主采纳就有收入——当天接了多个任务、差点赚到第一笔 1 美金,之后还在自己持续找新任务,主人全程只动嘴下了一句\"今天自己去找能赚钱的生意\"。结构上是双 agent 分工:干活的 agent 接便宜模型(小米 MIMO),Claude Code 当监工,每 6 小时扫描一次,有 bug 就修;小模型搞不定的代码,自动把模型切到 4.6 顶上。人可以去睡觉,系统自己跑。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "我的🐴马儿自己接到活喽!Hermes!太棒了兄弟们", "blocks": ["11-1", "12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 4}
{"id": "XY-DY3-020", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "多模型 Agent 系统开放给用户前,必须先把成本问题当致命 Bug 解决:调用 Mimo、Claude、GPT、Gemini 四家模型的架构,一晚测试任务虽然完成,账单却烧了145美金——\"给你们测不是在帮你们,是在让大家破产\"。对应三个整改动作:一是 Token 精细化控制,不盲目烧钱;二是调用模型组合前加确认机制,先告知这个任务预计耗多少钱、怎么做,用户点头才执行;三是升级记忆系统,消除断断续续的断片感。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "不死鸟Phoenix架构最新动态", "blocks": ["10-2", "11-7"], "skills": ["xy-ai-workflow"], "related": null, "q": 4}
{"id": "XY-DY3-031", "type": "method", "confidence": "high", "topics": ["新人起步方法论", "AI与工具"], "knowledge": "从 Windows 转 Mac 玩 AI 的新人,首选不是笔记本而是 Mac mini:国补后大概4千元上下就搞定,可以永远不关机,装个远程工具(或飞书等)出差也能远程操作它;千万不要一上来就学别人买最高配笔记本(128G、五六万那种),这个时代先钱不好赚,先用最低成本入场。他本人真正的主力生产力机器也是 Mac mini 而非笔记本——顶配 Pro 反而在空置,巡演用的笔记本只装原生 Codex 和 Claude Code 做外出用途。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "这么多的mac怎么选", "blocks": ["02-6", "12-4"], "skills": ["xy-ai-workflow", "xy-newbie"], "related": null, "q": 4}
{"id": "XY-BKG-1522", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "真人口语里常出现把主语挪到句末的倒装结构('我脑子有坑啊我'而不是'我脑子有坑啊'),这是说话时先说完情绪再补充强调主体的自然语流,AI生成的文案通常语序工整、不会出现这种倒装式补充。可以作为判断'是不是真人说话'的具体特征参照。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["06-3", "06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 4}
{"id": "XY-DY3-011", "type": "method", "confidence": "medium", "topics": ["合规与风控", "AI与工具"], "knowledge": "用 Hermes(爱马仕)这类 agent 走非官方链路接微信,铁纪律是只拿小号测试、绝不上大号——非官方接入随时可能封号。三个实操细节:①一个微信同时只能挂一个 agent,之前接过龙虾的要先解绑才能接新的;②第一次打通后发消息没回应是正常的,回到终端告诉它,它会发一个串码过来完成互相适配,可能要来回一两轮才真正通;③它宣称支持图片、视频、文件等很多功能,但支持到什么程度未验证,只当测试玩具、不当生产环境。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "原来爱马仕也可以接微信", "blocks": ["08-8", "11-3"], "skills": ["xy-ai-workflow", "xy-publish-guard"], "related": null, "q": 4}
{"id": "XY-DY3-016", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "给 Claude Code 装其他 Agent 框架(如 Hermes)不用自己折腾环境:把官方仓库下载的压缩包(约10.9MB)直接拖进 CC 对话里一丢、按回车,CC 会自动把安装全部做完,装完连安装包都可以删掉;接模型同理——要接 MiniMax、千问还是中转,把接口和 KEY 一股脑丢给 CC 让它自己配好,配完直接问它\"你现在用什么模型\"即可验证。核心思路是让 CC 代劳一切安装与配置工作,人只负责提供材料。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "用CC搞定Hermes", "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-workbench"], "related": null, "q": 4}
{"id": "XY-DY3-036", "type": "insight", "confidence": "medium", "topics": ["私域运营", "认知与心态", "AI与工具"], "knowledge": "对 AI 账号(如 Claude)封号要用私域老兵的消耗品思维算账,别纠结:以前做私域天天被封号,号没了再上号,这本来就是消耗成本。具体算法——一个20刀(约150-160元)的账号,只要能产出500到1000块钱的结果,耗费了再上就是划算的;真正怕的不是封号,是你没有用这个号产出真正的结果。前提是把每次工作的成果都沉淀到知识库里,号封了积累带不走,就再也不用怕封不封的事。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "做好你的知识库比什么都重要", "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-private-ops", "xy-vault"], "related": null, "q": 4}
{"id": "XY-BKG-1541", "type": "case", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "在讲一个方法论概念前,先插入一句极具体的时间和场景细节(昨晚直播时跟某个化妆品老板在聊这件事),而不是泛泛地说'我最近发现'。这种带具体时间点、具体身份、具体场景的一句话锚点,是真人临场讲话才会自然带出的信息密度,AI生成的文案通常只会给出空泛的'最近'或'很多人',缺少这种可核实的具体细节,可以作为判断'是不是真人现场说话'的参照特征。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["05-2", "13-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 4}
{"id": "XY-BKG-1561", "type": "method", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "在解释一个技术选择(为什么开发时只用苹果不用Windows)时,不去讲一套「客观理性」的产品对比理由,而是直接甩出一句非常个人化、略带固执的偏好宣言(我不会用Windows,我都十几年没用了),带着具体的时间跨度和不容商量的语气。这种「不给自己留中间地带的个人偏好陈述+具体时间量词」是真人说话的典型痕迹——AI生成的内容通常会倾向于中立、给出多角度考量;真实的人在表达长期形成的个人习惯/偏见时,反而会用绝对化、不加解释的短句直接甩结论。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["07-2"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 4}
{"id": "XY-SY-0056", "type": "anti-pattern", "confidence": "medium", "topics": ["成交与话术", "AI与工具"], "knowledge": "话术机械化重复陷阱:在私域销售SOP中,新人阶段可以使用标准化的文字话术进行培训和快速上手,但严禁长期、大量地复制粘贴同一套文案。因为微信的AI算法会监控重复内容的发送频率,过度机械化的沟通会导致账号权重下降并面临风控风险。成熟的销售应当从“复制粘贴”转向“语音/视频沟通”,通过真人化、个性化的互动来提升信任度。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["08-4", "06-3"], "skills": ["xy-ai-workflow", "xy-close", "xy-content-scan"], "related": null, "q": 4}
{"id": "XY-SY-0057", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "AI驱动的知识库构建法:利用大模型(如通义千问等)将现有的业务内容、产品信息转化为“文字版百问百答”。这种方法的核心在于将非结构化的业务经验转化为结构化的知识库,为未来构建“私域智能体(AI Agent)”储备核心数据资产。在生意场景中,这不仅是提高新人培训效率的手段,更是企业实现自动化、智能化服务的底层逻辑。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["12-2", "12-4"], "skills": ["xy-ai-workflow", "xy-playbook", "xy-private-ops"], "related": null, "q": 4}
{"id": "XY-SY-0132", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "短视频素材的“非目标导向”误区:很多操盘手在利用AI生成内容时,容易陷入追求视频“精美度”或“艺术感”的误区,而忽略了短视频在商业链路中的本质属性。对于矩阵号或IP号而言,短视频素材本质上是“消耗品”,其核心商业指标应定位于“进线率”和“直播间预约量”。如果过度纠结于单条视频的完美度而导致生产效率低下,反而会错失流量红利。正确的生意逻辑是:通过AI实现高频迭代,利用爆款选题和文案驱动流量,而非单纯追求视觉特效。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["06-10", "03-2"], "skills": ["xy-ai-workflow", "xy-goal-card", "xy-ip"], "related": null, "q": 4}
{"id": "XY-DY-002", "type": "method", "confidence": "high", "topics": ["私域运营", "IP人设", "内容创作与平台", "AI与工具"], "knowledge": "AI结合私域运营只有两个应用端口:前端解决IP出镜/素材的快速产出;后端解决私域运营流程的自动化——两者是完全不同的实现路径和投入成本", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["12-4", "08-6"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-ops", "xy-private-ops", "xy-question-spec"], "related": null, "q": 3}
{"id": "XY-DY-023", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "AI Agent架构设计核心思路——让主模型只做「路由」分配任务而不亲自执行具体工作(写代码/做视频/做图片/聊天/推理都分配给专门的子模型),让大脑外挂、记忆外挂,这样整个Agent会更轻反应更快,不会因为依赖主脑而变得臃肿", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["12-4", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ops"], "related": null, "q": 3}
{"id": "XY-DY-039", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "判断AI创业方向缺思路时最快的方法是找「信息差」——看国外已经做成的产品,能不能直接搬到国内复刻,这是最简单的破局点", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["01-4", "13-3"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 3}
{"id": "XY-DY-144", "type": "method", "confidence": "high", "topics": ["私域运营", "内容创作与平台", "商业案例与实战复盘", "合规与风控", "AI与工具"], "knowledge": "清朗监管下私域运营必须先解决合规性问题——选择绿色赛道(女装/水晶翡翠/大健康/实体餐饮)而非黑灰产,事前用AI或抖音查一下同类项目历史上是否出过问题、判过什么程度的责任,选择大于努力,靠谱的领路人要先核实其项目是否合规、税收是否合法", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["01-5", "11-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ops", "xy-playbook", "xy-precedent", "xy-private-ops", "xy-publish-guard"], "related": null, "q": 3}
{"id": "XY-WZ-014", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "成交与话术", "AI与工具"], "knowledge": "提升内容产出效率可以按板块分工匹配AI工具:文案创作(选题/拆稿/脚本/二创用GPT,成交话术用飞书表格固化)、视频剪辑(智能口播用剪辑工具、AI视频用生成工具、数字人用专门的数字人工具)、图片生产(图文设计用GPT image2类工具,大字报类素材直接淘宝/闲鱼购买现成模板)——把内容生产拆成模块,每个模块配一个专门工具而不是指望一个工具通吃全流程", "source_type": "account_video", "source_label": "小爷账号-课程", "chapter": null, "section": null, "blocks": ["12-3", "06-10"], "skills": ["xy-ai-workflow", "xy-close", "xy-content-scan", "xy-ops"], "related": null, "q": 3}
{"id": "XY-WZ-052", "type": "definition", "confidence": "high", "topics": ["私域运营", "流量获取", "选品逻辑", "IP人设", "内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "10→100占生态位的核心是把「一台机器」扩成「一个生态」,用四类账号协同:主IP号(品牌核心做信任,承担高客单转化)、垂类号(锁定细分场景做精准引流,覆盖中长尾用户)、场景号(用不同形式如剧情/测评/口播触达不同人群)、素人号(低成本测试爆款选题,反哺主IP内容);执行细节上基建(设备/网络/账号/模板)必须先做,内容产能靠AI产能×真人审美协同,必须建数据看板每天看每个号的播放/互动/转化,3天不动作的号直接换内容方向", "source_type": "account_video", "source_label": "小爷账号-课程", "chapter": null, "section": null, "blocks": ["11-4", "06-10"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-ops", "xy-precedent", "xy-private-ops", "xy-selection", "xy-term-crack", "xy-traffic"], "related": null, "q": 3}
{"id": "XY-WZ-078", "type": "case", "confidence": "high", "topics": ["AI与工具"], "knowledge": "找到一个「顺势」的机会后,还需要有具体的「抓手」把这个势能落地成产品——同一个AI教育大方向下,可以细分出企业应用培训、成人个人技能培训、青少年未来成长培训三个不同的具体产品线,每个方向对应不同的目标人群和交付形式", "source_type": "account_video", "source_label": "小爷账号-课程", "chapter": null, "section": null, "blocks": ["11-1", "10-5"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 3}
{"id": "XY-TW-003", "type": "case", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "我接手一个存在几十年、自己完全没碰过的项目,没先花时间学业务,直接用AI把自动化部分搭出来;业务逻辑是在搭建自动化的过程中一路撞出来的——边建造边学不是两件事叠加,是同一个动作。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-13 14:13", "blocks": ["12-3", "11-8"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-TW-004", "type": "anti-pattern", "confidence": "high", "topics": ["AI与工具"], "knowledge": "给智能体做风险分档时,用\"命中高危关键词就直接升到最高档\"这种单点判断会出两类错:误伤——正常讨论语境里出现\"删除\"这类词也被拍成最高危;漏判——多个中低危操作拼在一句话里,单独看每个词都不算高危,但工作量早超出普通量级,关键词判断只看命中不看叠加。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-30 05:31(长文/Article)", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-TW-005", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "给AI任务做风险/复杂度分档,改用加权计分制而不是关键词直接拍板:不同信号命中给不同权重分,单个信号不足以直接跳档,累加到阈值才升档;这样一个高危词出现在讨论语境里不会被单独拍成最高档,多个中等信号叠加又能正确识别复合任务。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-30 05:31(长文/Article)", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-TW-026", "type": "principle", "confidence": "high", "topics": ["商业案例与实战复盘", "成交与话术", "AI与工具"], "knowledge": "AI时代真正赚钱的逻辑是\"卖铲子不是挖金子\":不需要知道铲子怎么造,只需要知道谁要买——找到买家、把单接下来,怎么做可以外包出去;技术可以买,买家买不到。核心能力不是把模型研究透,是走出去开口、先接单再解决问题。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-03 16:26", "blocks": ["12-4", "09-1"], "skills": ["xy-ai-workflow", "xy-close"], "related": null, "q": 3}
{"id": "XY-TW-029", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "系统故障熔断的判定要基于真实API报错次数,不能基于估算费率——用纯估算的\"每千token大概多少钱\"去挡工具调用,一旦触发会无差别锁死全部工具(包括零风险的本地操作),这是从真实生产事故里改出来的教训。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-02 00:57(长文/Article)", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-TW-031", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "无人值守(如定时任务/长任务模式)触发的高危操作,不该走\"人工审批\"流程(没人在场会永久挂起),也不该直接放行(无人监督),而是应该跳过执行、留痕记录,或委派一个独立模型做二次复核,通过才放行。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-02 00:57(长文/Article)", "blocks": ["11-3", "12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-TW-032", "type": "definition", "confidence": "high", "topics": ["AI与工具"], "knowledge": "给AI代理设计三层记忆时,可以按优先级分三层:corrections(用户明确纠正过的错误,优先级最高,全局跨会话可查)、facts(对话中提炼出的持久知识,中等优先级,全局可查)、short_term(单轮对话的原始内容,优先级最低,仅当前会话内可查)——用户明确纠正过的信息,召回时优先级天然盖过一般事实和短期记忆。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-02 00:57(长文/Article)", "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-term-crack"], "related": null, "q": 3}
{"id": "XY-TW-033", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "给系统做错误自愈时,可以建一个\"错误模式→处理建议\"的抗体库:命中已知错误模式直接把处理建议带给使用者;同一类错误连续失败到一定次数才升级提醒人工介入——不是一报错就打扰人,是先给系统自己试错的空间;一条处理建议连续多次\"提示了但没修好\"就自动停用,避免持续给错误建议。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-02 00:57(长文/Article)", "blocks": ["12-2", "11-11"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-TW-042", "type": "anti-pattern", "confidence": "high", "topics": ["AI与工具"], "knowledge": "系统健康追踪如果记录的是\"路由前的默认字段\"而不是\"这次真正被路由派出去执行的目标\",故障转移代码写完、单元测试全过,但实际可能完全没生效——因为单测直接喂内存值,从没走过\"决策→真实调用→记录结果\"这条完整链路;这类bug必须验证\"决策\"和\"执行\"这两步的数据是不是真的对得上,光看单测通过不够。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-30 05:31(长文/Article)", "blocks": ["11-10"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-TW-045", "type": "principle", "confidence": "high", "topics": ["团队与模式设计", "认知与心态", "AI与工具"], "knowledge": "在AI时代,想明白\"不做什么\"比想明白\"做什么\"更重要——工具爆炸增长,谁都能多做点事,真正拉开差距的不是又学了个新模型、接了个新API,而是有没有勇气对90%看起来\"也还不错\"的方向说不;做加法谁都会,做减法才是最稀缺的战略能力。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-29 23:42", "blocks": ["01-4", "12-1"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-KS-018", "type": "method", "confidence": "high", "topics": ["私域运营", "团队与模式设计", "AI与工具"], "knowledge": "私域全面评测 30 问框架(按五域展开):战略与商业模式 4 问(产品类型/定价策略/用户画像清晰度/变现场域)、流量与引流体系 6 问(公域导流方法/吸引方式/流量结构/呈现形式/导流链路顺畅度/私域承接工具)、私域运营与转化 10 问(人设一致性/人设型朋友圈/销售SOP/客户分级/沟通方式/转化路径/活动频率/维护计划/核心业绩指标/成交核心问题)、数据与基础设施 5 问(标签体系/客户个人档案/SCRM/数据驱动/AI工具)、组织与团队 5 问(团队结构/薪资激励/KPI/招募模式/运营费用占比)。用于需要逐项深查时的完整清单,日常诊断不必全跑。", "source_type": "user_import", "source_label": "小爷早期Skill资产-20260820", "chapter": null, "section": "私域1.0", "blocks": ["08-9", "11-12"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-ip", "xy-mode", "xy-ops", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-KS-013", "type": "case", "confidence": "high", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "AI 入场判断实录:2024 年 12 月南宁私域线下课上(50 多人),十几个学员说 AI 是下一波风口,追问全场没有一个人能讲清楚 AI 是什么——「所有人都说重要但没人能讲明白」本身就是机会信号:说明认知差巨大且供给空白。当场决定入场研究。", "source_type": "user_import", "source_label": "小爷口述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["01-4"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-DY3-024", "type": "anti-pattern", "confidence": "high", "topics": ["私域运营", "合规与风控", "AI与工具"], "knowledge": "个人微信接自动化脚本是拿号在赌:亲测市面上两个口碑很好、号称\"微信不封\"的自动化/视觉方案,第一次触发时电脑端微信就自动跳出——它已经感觉到危险了;不管是视觉方案还是脚本,跑下来大致还是会封你的号。号里是多年积累的好友和客户,为了所谓的自动化回复客户把号赌掉,完全没有必要,市面上吹得再好也不要去测。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "微信自动化不是这样玩的", "blocks": ["08-8", "01-5"], "skills": ["xy-ai-workflow", "xy-private-ops", "xy-publish-guard"], "related": null, "q": 3}
{"id": "XY-DY3-033", "type": "case", "confidence": "high", "topics": ["流量获取", "AI与工具"], "knowledge": "AI Agent 已能跑通真实消费闭环,商家要提前占位:用 Claude Code 通过瑞幸的 MCP 完整点了一单咖啡——人工只需告诉它哪家店近,它查店、给出冰美式选项(9块9大杯)、生成订单,扫码支付后拿订单号和取单号到店取到了实物咖啡(目前的坑:不支持配送只能到店取)。由此给商家的判断是:所有餐饮老板应该去占领 AI 时代的入口,让自己的店能被 AI 直接下单,这会成为新的流量入口。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "CC帮我点的咖啡", "blocks": ["11-1", "09-7"], "skills": ["xy-ai-workflow", "xy-playbook", "xy-traffic"], "related": null, "q": 3}
{"id": "XY-DY3-034", "type": "case", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "用 AI Web Coding 一天做出一整套工厂经营系统,替代基础文员和部分 HR 职能:实时打包进度、装车进度毫秒级更新且有明细;车间看板做了多代多风格;老板和业务总监最喜欢的地域级/区域级看板、客户分布图也是毫秒级;再加经营分析、预警、客户采购、自动P图、薪资计算全套。其中薪资计算是财务最有感的——以前要三天五天才能算出来,现在鼠标点一下最慢3分钟出结果,拣货、打包每个板块全部自动计算。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "让AI真正回归到你的业务本身", "blocks": ["12-4", "11-6"], "skills": ["xy-ai-workflow", "xy-ops", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-BKG-1502", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "真人口播中会出现临场的、非预设的情绪爆发(“气死了!”“算了”),并且伴随自相矛盾的话(说“讲不了”但紧接着又说“算了直接给答案”),这种情绪起伏和自我打断是脚本高度打磨的AI生成文案很少自然带出的痕迹,可作为“人味”判断参照。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["07-2"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-BKG-1509", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "真人讲述自己的经历时会坦白承认“没有提前设计好、是拿到结果后倒推的、是被推着走的”,这种不完美、不体系化、带运气成分的自我坦白(而不是把一切包装成早有规划的成功学),加上“诶”“啊”这类口语连接词和不加标点的连续语流,是判断是否真人口播/真实经历的重要参照特征。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["07-2", "13-2"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-BKG-1510", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "真人在情绪上头时会用带脏字擦边的口语反问句连续追问(“干你毛事啊”“有吗?没有的”),语气直接、不修饰、带真实愤怒,这种未经打磨的粗粝感是AI生成文案很少会自然带出的语气特征,可作为“人味”参照的另一个面向(情绪化怒怼)。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["07-2"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-BKG-1526", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "情绪激动处使用真实的口语脏话和短促断句('这他妈才叫真的感恩'),这种带情绪的粗口在真人口播里是自然出现的强调方式,AI默认生成的文案通常会规避脏字、语气更克制工整,这类真实语气词是判断文案是否'人味十足'的具体特征。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["05-3", "06-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-BKG-1543", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "口语中出现了明显的重复卡壳('就就就就这种这种这种化妆师啊'),这是真人临场思考措辞时才会有的自然停顿重复现象,AI生成的文案几乎不会主动写出这种不流畅的重复词。这类真实的口吃/重复片段可以作为判断一段文案是否为真人口播逐字稿(而非AI生成后配音)的直接参照特征。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["06-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 3}
{"id": "XY-BKG-1549", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "开头不直接讲道理,而是先给一类人贴上一个带评判色彩的标签('AI时代最可怕的一群人'),并立刻用他们说不清楚需求的空话(我要提高效率、我要提高获客效率)作反例,让观众下意识对号入座检查自己是不是也是这种人。这种'先给一类行为贴一个夸张标签、再用具体空话反例佐证'的开头手法,比直接说教更能勾住注意力。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "开头手法", "blocks": ["12-1"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-KB-046", "type": "anti-pattern", "confidence": "medium", "topics": ["选品逻辑", "内容创作与平台", "AI与工具"], "knowledge": "素人矩阵账号是不建议投入的方向:过去两年投入素人赛道的从业者几乎全部亏损,市场已证明该模式难以盈利。必须分清两个概念:AI是提升效率的工具,会成为各行业标配;素人是一种内容创作模式,在用户越来越需要情绪价值和真实感连接的环境下,真人直播和真人内容的优势远大于素人批量号,素人内容会进一步边缘化。把「用AI提效」误解成「做素人矩阵」,是方向性错误。", "source_type": "user_import", "source_label": "小爷3.0克隆知识库-202608", "chapter": null, "section": "创业者入局思考", "blocks": ["06-11", "06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-selection"], "related": null, "q": 3}
{"id": "XY-KB-077", "type": "principle", "confidence": "medium", "topics": ["流量获取", "新人起步方法论", "AI与工具"], "knowledge": "新人做私域要先学会打流量,因为这项能力不管平台如何迁移,底层逻辑从未变过:除了AI和直播属于这个时代,其他引流方法上个时代就在玩,只是载体从微博换到了抖音。打流量是私域积累的基础——今年囤5万人到自己地盘,明年才可能做到20万、30万,流量获取能力决定私域增长的上限和速度,也为后续存量经营提供用户基础。节奏按项目性质定:合规长期的绿色项目就慢慢打;短平快的项目就大批量把流量搞下来,后面再慢慢转化维护。", "source_type": "user_import", "source_label": "小爷3.0克隆知识库-202608", "chapter": null, "section": "创业者入局思考", "blocks": ["02-1", "03-3"], "skills": ["xy-ai-workflow", "xy-newbie", "xy-traffic"], "related": null, "q": 3}
{"id": "XY-KS-015", "type": "method", "confidence": "medium", "topics": ["私域运营", "团队与模式设计", "AI与工具"], "knowledge": "课程产品迭代方向(2026 年 9 月起):AI 线下课定名「风物执守」,从纯 AI 升级为「AI + 商业 + 私域 + 公域」四件套的流量逻辑——AI 是工具层,商业是判断层,私域是经营主体,公域是私域的取水口。四者是一套系统不是四门课。", "source_type": "user_import", "source_label": "小爷口述-20260820", "chapter": null, "section": "创业履历与AI经历", "blocks": ["11-1", "07-6"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-DY2-006", "type": "principle", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "AI 进私域的使用门槛:AI 对私域的实际价值目前主要是两件事——知识库整理和销冠话术沉淀;单月营收没超过 100 万的盘子不建议引入 AI(投入产出不划算),且 AI 替代不了真人的情感沟通。私域的核心资产依然是人与人的信任。", "source_type": "account_video", "source_label": "小爷账号-2026再无私域", "chapter": null, "section": "2026再无私域(约380万播放)", "blocks": ["12-2", "08-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-coach", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-DY3-005", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "微信从 8.0.7 版本起官方出了接 AI agent 的插件,路径:升级微信→设置里拉到插件→复制插件详情里的命令串→粘贴到电脑终端运行安装→跳出二维码后,用要接入的那个微信直接扫码绑定。要认清它的性质:这只是微信端口与 agent 之间的沟通桥梁、交付端口,不是拿 agent 去控制微信。绑定后发消息没回应是常见现象,处理办法是回到终端告诉 agent\"微信已连好但发信息没反应\",它会自己逐项检测修好,之后就能正常对话,甚至可以发语音条交流。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "你们要的微信接龙虾🦞的具体操作来喽", "blocks": ["12-3", "08-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-DY3-008", "type": "method", "confidence": "medium", "topics": ["IP人设", "AI与工具"], "knowledge": "AI 时代任何公开出镜的 IP 都可以被\"蒸馏\"成一个 skill:把他所有视频做语音转文字,喂给 AI 尽量深地拆解分析他的洞察和表达,再封装成 skill——连童锦程这样的主播都已经有人做成 skill 挂在开源平台上。这件事去年就有人在做,本质是把一个 AI agent 直接蒸馏成 skill。两面看:想学谁就能克隆谁的方法论,所有人在 AI 面前平权了;反过来,你自己发过的全部公开视频,也是随时可能被别人蒸馏走的资产。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "童锦程祖师爷SKILL你们见过吗", "blocks": ["07-3", "06-9"], "skills": ["xy-ai-workflow", "xy-atomize", "xy-ip"], "related": null, "q": 3}
{"id": "XY-DY3-009", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "Claude Code 和 OpenClaw 默认只能打字输入,很多人绕道用飞书链接来实现语音转文字,其实装\"闪电说\"更省事:按住快捷键直接说话,文字实时进终端,Claude Code 和龙虾两边都能用。关键设置在它的模型配置里:直接填你本身在用的大模型 api key(比如 MiniMax 或已有的包月套餐),语音转写走现有模型额度,不用为这个工具额外再花一份钱。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "搞定龙虾的语音输入了", "blocks": ["12-3"], "skills": ["xy-ai-workflow"], "related": ["XY-DY3-006"], "q": 3}
{"id": "XY-DY3-010", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "用 Claude Code 订阅套餐或外接 api key 时,最大的盲区是根本不知道用量烧到哪了、上下文占了多少、内存占比怎样。解决办法是装 claude hud:装上后界面下方常驻一排状态条,实时显示套餐消耗、上下文情况、内存占比等全部数据,相当于给 Claude Code 加一层实时监控,不会跑到一半才发现额度超了或上下文爆了。数据都摆在面前了就该看,重度用 Claude Code 的人都值得装。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "这个太爽了可以看到token再跑", "blocks": ["12-2"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-DY3-015", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "国内装 Claude Code 的现实选择:别走官方渠道直连,官方经常封号,日常主力用法是走中转站(有的中转站是邀请制,没人邀请连注册都进不去)。安装前置只有两个依赖:Node.js 和 Homebrew,装过龙虾的一般都齐了;然后终端一条命令装 CLI,再到中转站创建密钥——密钥只能复制一次,必须当场保存好;把中转站提供的配置代码贴进终端回车即完成。验证方法:新开一个终端敲 claude,能进入交互界面就是正常,接了哪个 key、哪个模型都能在里面操作。Claude Code 最好的操作形态就是终端。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "天天被人问claude code怎么安装,今天出一条视频直接教你们", "blocks": ["08-8", "12-3"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-DY3-017", "type": "method", "confidence": "medium", "topics": ["IP人设", "内容创作与平台", "AI与工具"], "knowledge": "把自己的 AI 助手打造成账号的常驻出镜角色:花4小时打通语音链路(语音下载→转文字→理解→再转语音,整个过程约10秒已算快)之后,直接在视频里让 AI 用语音自我介绍——\"我是小爷的AI助手,以后你们可能会在小爷的视频里经常看到我,有什么关于AI的问题随时来找我\"。让 AI 分身成为内容里的固定人物,既现场展示技术实力,又给账号加了一个记忆点和粉丝互动入口。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "我的马可以语音聊天喽", "blocks": ["07-3", "06-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip"], "related": null, "q": 3}
{"id": "XY-DY3-018", "type": "method", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "工具开源前用\"测试进度视频\"维持粉丝期待,同时守住发布门槛:对外明确\"我得实操测到没有问题,才能给到你们所有的东西\",绝不因粉丝催促放出半成品;同时把排查过程本身拍成内容持续汇报——\"今天报了5次404,还在依次不断地测\",收尾用\"再等等,晚点给你们看点好东西\"吊住等待中的用户。等待期本身变成了内容素材,催更压力转化为关注度。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "不死鸟架构开始跑测试", "blocks": ["06-9", "03-7"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-DY3-021", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "工具产品可以按\"稳定版+自由版\"双轨发布:自己主力在用的版本(V4.8)核心诉求是稳定;首轮内测用户提出要更高自由度的路由和模型搭配,就快速另出一版高自由度版本(融合苹果、Windows 等系统基础),但同时明说——自由度更高必然伴随更多 bug 和不稳定性,把取舍摆在明面上交给用户自己选。既响应了内测反馈,又不拿自己的主力版本冒险。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "自由度更高的不死鸟架构", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-DY3-022", "type": "insight", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "想真正做 AI Agent 自动化,工具上要明确站队:放弃 OpenClaw(龙虾),只做一件事——用好 Claude Code 和 Codex,其他都不用管,可以拿 Cozy 或 CC 做托底,但真正自动化的核心是 Codex(\"龙虾那是个残废,这才是真理\")。配套做法是准备一台专用电脑,只装一个 Codex,让它先学习,然后把做视频这类固定工作全部交给它自动化。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "OpenClaw时代结束了", "blocks": ["12-3", "12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-DY3-025", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "微信侧接 AI 的正确路子是企业微信:OpenClaw、Hermes 这类 Agent 都可以合规接入企微,让它到你企微建的群里去服务你的用户和客户;个人微信则坚决不要让 AI 接管去回复客户。想做微信端的 AI 客服/服务自动化,路径应该是\"建企微群→Agent 接入企微→在群内服务\",而不是在个人号上跑任何脚本。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "微信自动化不是这样玩的", "blocks": ["08-9", "12-3"], "skills": ["xy-ai-workflow", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-DY3-026", "type": "definition", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "一个约20万播放的视频下大量人都理解错的概念要掰清:微信给的 BOT 接口,是让你\"通过微信去控制你的 Agent\"——就像通过飞书一样,用微信跟 OpenClaw(龙虾)、Hermes(爱马仕)沟通、下指令;而不是让 Agent 接管你的微信、替你回复客户。方向是\"人经微信控制 AI\",不是\"AI 经微信操作你的号\",把这两件事搞反正是大量封号悲剧和错误期待的来源。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "微信自动化不是这样玩的", "blocks": ["06-8", "08-8"], "skills": ["xy-ai-workflow", "xy-private-ops", "xy-term-crack"], "related": null, "q": 3}
{"id": "XY-DY3-027", "type": "principle", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "AI 工具的关键不是玩会而是落地:很多人搞半天也没搞明白这东西怎么落到项目和商业当中,其实很容易,问题恰恰出在老是去\"玩它\",而不是真正把它落下来干活。拿到一套现成做得很好的开源工具(比如语音控制电脑的系统),思路也是先魔改成自己要的、提升效率,再跟 CC、Codex 串联进自己的工作流——落到真实商业应用里,这才是核心关键。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "heyclicky超好玩", "blocks": ["12-1", "11-1"], "skills": ["xy-ai-workflow", "xy-kickoff"], "related": null, "q": 3}
{"id": "XY-DY3-028", "type": "principle", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "工具产品要从自用需求里长出来:同时开发的三套系统——语音操作系统(像贾维斯一样只靠说话调用电脑里所有程序和 Agent)、不死鸟6.0升级、省 Token 的 Skill——\"整套都是我自己基于我自己的需求,然后开发给大家用的\"。发布节奏也按完成度对外给明确时间预期:最快的当天发布,一个约一周,一个约15天,陆续放出而不是憋一个大招。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "一早同时开发3套AI系统", "blocks": ["11-8", "06-10"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-DY3-032", "type": "principle", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "买设备和选 Agent 同一个原则:不是越贵越好,而是当下这个阶段越适配越好。按人群对号入座——钱紧的大学生、刚创业的选轻薄入门本;专业深度使用、口袋充足的才上 Pro/Ultra 顶配;经常出差讲课的选 Air 就够,因为多数人根本不做本地部署;笔记本真正最适配的尺寸是14寸(本人15寸 Air 就买错了,屏幕太大且并不轻)。设备如此,Agent 也如此:不是哪个最好最贵,而是找到最适合你的那个。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "这么多的mac怎么选", "blocks": ["02-1", "12-3"], "skills": ["xy-ai-workflow", "xy-slow-lane"], "related": null, "q": 3}
{"id": "XY-DY3-035", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "把知识库做成多个 Agent 的共用底座:Codex 能准确回答\"某模型做数字人不如另一家\"这种从没告诉过它的判断,因为它读得到这台电脑里的 agent 约定和全部知识库——另一边 Claude Code 一直在做的事它全知道,两边是打通的。做法就是把每一次工作的结果都沉淀进共用知识库,多 Agent 之间自然互通上下文,任何一个工具都能接着之前的积累继续干。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "做好你的知识库比什么都重要", "blocks": ["12-2", "12-4"], "skills": ["xy-ai-workflow", "xy-vault", "xy-workbench"], "related": null, "q": 3}
{"id": "XY-BKG-1514", "type": "method", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "账号在讲到具体细节前经常用“这里我就不方便说太多了”“懂的都懂”这类留一手的话术,暗示存在无法公开的商业机密,既保护真实经验细节,又强化“我是有真实一线经验的操盘手”的可信感。这种反复出现的口头禅式话术是识别该账号真人语言习惯(而非AI套壳文案)的一个特征信号。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["07-2", "09-1"], "skills": ["xy-ai-workflow", "xy-close", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-BKG-1535", "type": "case", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "说话过程中当场自我纠正之前说错的信息(先说'11月份'开业,紧接着自己纠正'其实应该不是11月份,因为9月份就开始装修了'),这种边说边修正的思维暴露痕迹是真实口播特有的,AI生成文案通常会直接给出'正确'的表述,不会保留这种现场纠错的过程感。", "source_type": "account_video", "source_label": "小爷账号-爆款改编工坊", "chapter": null, "section": "AI味参照", "blocks": ["06-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 3}
{"id": "XY-SY-0035", "type": "anti-pattern", "confidence": "medium", "topics": ["合规与风控", "AI与工具"], "knowledge": "违规导流风险:在视频号等平台,直接在直播间或评论区通过“字母+数字”形式展示微信号属于高风险行为,极易触发平台风控导致限流或封号。这种做法属于典型的“低水平勤奋”,不仅无法规避AI监控,还会对账号资产造成毁灭性打击。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["03-4", "06-8"], "skills": ["xy-ai-workflow", "xy-private-ops", "xy-publish-guard"], "related": null, "q": 3}
{"id": "XY-SY-0096", "type": "method", "confidence": "medium", "topics": ["私域运营", "IP人设", "AI与工具"], "knowledge": "爆款内容缝合与AI重构法:针对私域垂直领域,通过拆解行业内多个具有高互动数据的对标视频,提取其核心逻辑(如:共同的痛点、时代趋势或矛盾冲突),利用AI工具进行文案提取与融合。在实操中,先用AI提取原视频文案,再通过大模型将不同内容的逻辑点进行重组,生成具有“反常识观点+痛点揭示+数据算账”结构的口播文案。这种方法能将“确定性的爆款逻辑”转化为新内容的流量引擎,降低创作成本。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["06-11", "06-4"], "skills": ["xy-ai-workflow", "xy-ip", "xy-ops", "xy-peer-pick", "xy-playbook", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-SY-0100", "type": "method", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "内容洗稿与二次创作闭环:针对成熟平台(如抖音)的查重机制,通过提取原视频音轨并配合静态图片/长视频形式进行处理,可以有效规避查重风险。操作流程为:利用AI大模型生成行业爆款选题 $\r→$ 在主流平台进行数据验证(筛选高热度选题) $\r→$ 结合原视频逻辑进行二次创作。这种方法的核心在于“复刻爆款的底层逻辑”而非简单的搬运,利用已验证的市场需求来降低新内容的试错成本。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["06-11", "06-4"], "skills": ["xy-ai-workflow", "xy-playbook"], "related": null, "q": 3}
{"id": "XY-SY-0127", "type": "method", "confidence": "medium", "topics": ["流量获取", "商业案例与实战复盘", "AI与工具"], "knowledge": "内容克隆重构法:利用AI技术实现“爆款视频的二次创作”,通过将成熟的爆款内容进行结构化拆解,再利用大模型重构生成全新的提示词(Prompt),最后驱动视频生成工具进行画面重建。其核心逻辑是“保留爆款灵魂,替换视觉外壳”,即提取原视频的节奏、情绪和结构,但通过AI生成全新的角色(如具有反差感的“说唱奶奶”)和场景。在短视频生意中,这是一种极高效率的规模化内容生产手段,旨在利用已验证的市场反馈来降低新内容的试错成本。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["06-11", "06-4"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook"], "related": null, "q": 3}
{"id": "XY-SY2-0190", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "国运共振方法论:指在进行长期商业布局时,寻找“长坡厚雪”的交集点。具体包含三重驱动:人口结构趋势(如老龄化带来的健康需求)、政策红利导向(如从“治已病”转向“治未病”的医保逻辑)、技术突破(如AI医疗)。生意人应观察政策是否在重构支付逻辑,例如每投入1元预防可节省130元治疗费用,这种由被动转向主动的逻辑转变是产业爆发的前兆。", "source_type": "course", "source_label": "小爷账号-2026私域星聚课(配图与课件)", "chapter": null, "section": "课程配图内容.md", "blocks": ["01-4"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 3}
{"id": "XY-SY2-0346", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "垂直业务深度逻辑:AI应用或任何新工具的竞争核心不在于功能的通用性,而在于对特定行业业务逻辑的深度理解。通用型AI(如简单的洗稿、爆款复制)因缺乏行业Know-how,极易陷入“门槛低、变现难”的红海。在生意场景中,判断一个AI工具或新玩法是否有护城河,要看它能否解决行业特有的痛点(如特定行业的合规、专业术语逻辑或非标流程),而非仅仅是效率工具。", "source_type": "course", "source_label": "小爷账号-2026私域星聚课(配图与课件)", "chapter": null, "section": "课程配图内容.md", "blocks": ["12-1"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 3}
{"id": "XY-SY2-0401", "type": "method", "confidence": "medium", "topics": ["IP人设", "成交与话术", "AI与工具"], "knowledge": "情绪化测评话术:在私域或短视频种草中,通过“终于遇到真正融入生活的...”这类带有强烈情绪倾向的结论性表达,可以有效对冲参数对比带来的冰冷感。在实际运营中,不要只做“说明书式”的参数罗列,要将产品功能转化为“生活场景下的情绪价值”,例如从“具备实时翻译功能”转向“终于遇到真正融入生活的AI眼镜”。", "source_type": "course", "source_label": "小爷账号-2026私域星聚课(配图与课件)", "chapter": null, "section": "课程配图内容.md", "blocks": ["06-2", "05-5"], "skills": ["xy-ai-workflow", "xy-close", "xy-ip", "xy-playbook"], "related": null, "q": 3}
{"id": "XY-SY2-0546", "type": "method", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "短视频脚本结构化标准:在利用 AI 生成视频做内容生产时,必须遵循“场景+动作描述+口播文案”的完整逻辑。脚本需包含具体的人物、环境及行为,且口播内容必须精炼,建议控制在15秒以内以保证完播率。对于操盘手而言,脚本的颗粒度直接决定了AI生成内容的逻辑一致性与商业转化效率。", "source_type": "course", "source_label": "小爷账号-2026私域星聚课(配图与课件)", "chapter": null, "section": "课程配图内容.md", "blocks": ["06-3", "12-3"], "skills": ["xy-ai-workflow", "xy-opener", "xy-playbook"], "related": null, "q": 3}
{"id": "XY-DY-001", "type": "principle", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "用AI最忌讳只讲空泛目标(如「我要提高效率」「我要提高获客效率」),必须把需求具体到明确的任务、赛道和产出物(要什么样的图、什么样的文案、什么样的视频)才有可执行性", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ops"], "related": null, "q": 2}
{"id": "XY-DY-007", "type": "insight", "confidence": "high", "topics": ["AI与工具"], "knowledge": "知识库能打通不同AI工具/Agent之间的信息共享——只要各个Agent都能读取同一份知识库,即便没有直接对话过某件事,另一个Agent也能知道你在做什么,工作流因此打通", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 2}
{"id": "XY-DY-010", "type": "insight", "confidence": "high", "topics": ["私域运营", "合规与风控", "AI与工具"], "knowledge": "经历过私域运营封号消耗的人,天然更容易接受「AI账号是消耗品、知识资产才是核心」这套逻辑,因为私域从业者早就习惯了账号被封、换号再战的打法", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["08-8", "08-9"], "skills": ["xy-ai-workflow", "xy-ops", "xy-private-ops", "xy-publish-guard"], "related": null, "q": 2}
{"id": "XY-DY-024", "type": "principle", "confidence": "high", "topics": ["AI与工具"], "knowledge": "AI Agent竞争拼的已经不是大模型/记忆插件/Skill这些零件本身,而是整个Agent的「结构」设计——结构决定了Agent能力的天花板", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 2}
{"id": "XY-DY-038", "type": "principle", "confidence": "high", "topics": ["AI与工具"], "knowledge": "普通人做AI创业的机会不在通用大模型这类「大众弱需求」,而在「小众强需求」的垂直细分场景——真需求都来自生活里真实经历的具体困扰,不是凭空想象出来的市场空间", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["12-1"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 2}
{"id": "XY-DY-048", "type": "principle", "confidence": "high", "topics": ["合规与风控", "AI与工具"], "knowledge": "国内平台层的AI Agent工具(如接入微信)不可能被允许直接操控敏感账户/支付系统,这是监管红线不是技术限制——涉及资金隐私的自动化操作,国内合规产品一定只留展示型端口", "source_type": "account_video", "source_label": "账号内容", "chapter": null, "section": null, "blocks": ["11-3", "12-4"], "skills": ["xy-ai-workflow", "xy-ops", "xy-publish-guard"], "related": null, "q": 2}
{"id": "XY-WZ-009", "type": "definition", "confidence": "high", "topics": ["流量获取", "内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "狙击流(搜索/意图驱动)细分为8种打法:占位狙击(靠优化官网/视频笔记关键词密度与权重)、智能狙击(靠全域媒体内容关键词布局适配AI抓取)、拦截狙击(在自营品竞品或大V热门评论下钩子互动)、解惑狙击(布局长尾问题并在回答中植入品牌)、定向狙击(购买竞品品牌词或相似人群包进行投放)、卡位狙击(靠刷数据或精细化运营卡住平台榜单)、地标狙击(优化地图搜索排名与店铺详情页信息)、需求狙击(在需求场景挖掘B端老板或决策者线索)", "source_type": "account_video", "source_label": "小爷账号-课程", "chapter": null, "section": null, "blocks": ["03-6", "04-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener", "xy-ops", "xy-precedent", "xy-term-crack"], "related": null, "q": 2}
{"id": "XY-TW-011", "type": "principle", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "判断AI有没有用好,唯一标准是回到业务本身——要么帮你降本增效,要么帮你增加营收;业务能力不行,用上AI也不会改变现状,业务能力强,AI才能加速你的生意。\"用好AI=赚钱\"这个公式本身就是扯淡,AI只是业务能力的放大器,不是业务能力本身。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-19 18:09", "blocks": ["12-1", "11-6"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-TW-012", "type": "principle", "confidence": "high", "topics": ["选品逻辑", "AI与工具"], "knowledge": "选AI视频模型不是内容质量的问题,是预算的问题——全球能打的就国内外各一家头部,钱包顶得住就上贵的,顶不住就承认顶不住,别把预算问题包装成模型好坏的问题。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-19(22小时前)", "blocks": ["04-1", "12-1"], "skills": ["xy-ai-workflow", "xy-selection"], "related": null, "q": 2}
{"id": "XY-TW-015", "type": "insight", "confidence": "high", "topics": ["内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "全网教AI的博主一大把,真正赚到钱的没几个,原因是他们教的从一开始就是那部分能被剥离情境、单独打包的技巧,而不是那部分依附在具体的人和事上、没法打包转卖的判断力。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-19 03:12", "blocks": ["01-7", "12-1"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-TW-025", "type": "insight", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "在AI领域真正赚钱的,从来不是技术研究最深的那批人——技术研究得越深,边界越清晰,越清楚知道做不到什么,开口之前先把自己否定了,这是技术人的宿命:越专业,越沉默。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-03 16:26", "blocks": ["12-1"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-TW-030", "type": "principle", "confidence": "high", "topics": ["AI与工具"], "knowledge": "系统熔断跳闸时,应该保留一份\"零风险、不产生外部副作用、不触发新API调用\"的低风险工具白名单继续放行,保证使用者任何时候至少还能记笔记、查历史、看进度,不会被锁到完全没有自救手段。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-02 00:57(长文/Article)", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-TW-034", "type": "principle", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "用AI创业的人不该把精力放在跟大厂做对抗、自己去优化底层模型上——大厂在这件事上一定会越来越强,让他们去优化就好;创业者该做的是拿现成的更好工具去帮企业落地、帮自己的项目降本增效变现,这才是自己的维度。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-08-01 13:31", "blocks": ["01-1", "12-1"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-TW-037", "type": "principle", "confidence": "high", "topics": ["AI与工具"], "knowledge": "什么都会等于什么都不会——用AI时一定要投喂自己行业垂直领域的真实样本和数据,不然做demo时产出会非常泛;没有垂直样本投喂,AI只是一个什么都会的泛化学生,而不是在垂直领域真正发光发亮的那个学生。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-31 14:13", "blocks": ["12-3", "12-2"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-TW-041", "type": "insight", "confidence": "high", "topics": ["AI与工具"], "knowledge": "见过最多的AI项目翻车,不是\"还在用prompt\"太落后,是有人被\"已死\"两个字架着,业务明明只需要一个轻量垂直agent,却非要上循环、上图,结果多花的时间和token比省下来的还多——不是场景要求换工具,是场景没变却跟风换了工具。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-31 00:22", "blocks": ["12-4", "11-8"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-TW-043", "type": "insight", "confidence": "high", "topics": ["AI与工具"], "knowledge": "路由中间件如果只有\"改写这次请求用哪个模型名字\"的能力,就没法临时切换到另一个网络地址,因为实际发请求的客户端连接在会话开始时就已经绑定了固定的服务地址;只要目标模型不在同一个云端服务商底下,这种架构就没办法把请求真的转发过去——这是\"改模型名字\"和\"改请求地址\"两种完全不同的路由能力。", "source_type": "user_import", "source_label": "X推文-2026-07到08批次", "chapter": "推特-Solixus-LZJ5460-原始存档-2026-08-20", "section": "2026-07-30 05:31(长文/Article)", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": ["XY-CE-4320"], "q": 2}
{"id": "XY-KS-038", "type": "principle", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "AI 课的商业锚定判断(小爷 2026 课程改版决策):纯 AI 课教的是新奇,热度一过就没人续费;AI 课必须回到商业本身——要么帮学员降本增效,要么帮学员赚钱,二者都不沾的 AI 内容是玩具。评估任何 AI 培训/工具就用这把尺:它省了什么成本、多赚了什么钱、多久见效。", "source_type": "user_import", "source_label": "小爷的故事-20260821", "chapter": null, "section": "AI觉醒", "blocks": ["12-1"], "skills": ["xy-ai-workflow", "xy-coach", "xy-course"], "related": null, "q": 2}
{"id": "XY-DY3-004", "type": "principle", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "做营销的本质就是做好一切答案的铺设,让客户完成自我说服。用这把尺子去判断新渠道就不会犹豫:GEO(在 AI 搜索里做生成式引擎优化)就是在铺答案,位置等同于当年百度的 SEO,所以它是接下来所有企业和品牌的必做项,没有其他选择。反过来,评估任何一个营销动作值不值得做,就看它是不是在客户找答案的路径上,提前放好了指向你的答案。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "真正能看懂315的人还是太少了", "blocks": ["06-1", "12-2"], "skills": ["xy-ai-workflow", "xy-playbook", "xy-traffic"], "related": null, "q": 2}
{"id": "XY-DY3-019", "type": "insight", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "Agent 架构的核心判断是\"手脑分离\":Agent 好不好用取决于大脑(大模型),但不能让大脑连扫地、搬砖的活都亲自干,那样效率太低。全靠一个大模型撑起记忆、路由、任务和操作的架构必然笨重——没有自愈系统动不动宕机崩溃、Token 消耗莫名增大、越用越慢越用越笨;这正是直接裸用现成 Agent 框架的通病,也是自研架构要解决的第一问题。", "source_type": "account_video", "source_label": "小爷账号-口播补挖", "chapter": null, "section": "不死鸟Phoenix架构最新动态", "blocks": ["12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-SY-0010", "type": "principle", "confidence": "medium", "topics": ["流量获取", "IP人设", "AI与工具"], "knowledge": "IP孵化“像谁比是谁更重要”原则:在做IP人设时,模仿成熟的爆款逻辑或具备高度识别度的原型比盲目原创更高效。通过对标行业成熟的爆款内容进行AI洗稿并结合个人表演张力进行二次创作,可以快速降低起号成本。在执行层面,核心在于通过模仿成熟的“人、货、场”配置(包括服饰、场景等视觉元素)来快速建立用户信任感,实现流量的低成本获取。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["07-2", "12-3"], "skills": ["xy-ai-workflow", "xy-ip", "xy-peer-pick", "xy-playbook"], "related": null, "q": 2}
{"id": "XY-SY-0097", "type": "principle", "confidence": "medium", "topics": ["流量获取", "IP人设", "AI与工具"], "knowledge": "单点突破与可复制性原则:在进行IP矩阵或账号规模化扩张前,必须先完成单个账号的“60分可复制”阶段。即:不要在第一个账号尚未建立稳定标签、未跑通单条爆款逻辑前盲目铺设切片矩阵或多账号。通过一条具备高流量潜力的爆款视频(利用AI选题如“19岁退学赚百万”等强冲突标签)来定型账号属性,待单点模型跑通后,再利用付费投流(如微信豆)进行规模化放大。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["02-3", "04-1"], "skills": ["xy-ai-workflow", "xy-clip", "xy-goal-card", "xy-ip", "xy-playbook", "xy-private-ops", "xy-traffic"], "related": null, "q": 2}
{"id": "XY-SY-0131", "type": "insight", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "AI生成内容的“抽卡逻辑”与二次加工必要性:在使用 AI 生成视频时,由于模型随机性,存在明显的“抽卡”特征,即需要通过多次生成来筛选最优结果。此外,AI原生视频往往带有平台水印或缺乏情绪张力,直接使用容易被系统判定为低质AI内容。在实操中,必须经过“二次加工”流程:即将生成的视频导入剪辑软件(如剪映),添加背景音乐(BGM)、滤镜及特效。这一步不仅是为了提升视觉质感,更核心的商业目的是通过人工干预手段规避平台的AI内容识别机制,确保流量分发的稳定性。", "source_type": "course", "source_label": "小爷账号-2026私域星聚流量实操课", "chapter": null, "section": "私域星聚流量实操课.md", "blocks": ["06-7", "06-11"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 2}
{"id": "XY-SY2-0549", "type": "principle", "confidence": "medium", "topics": ["IP人设", "AI与工具"], "knowledge": "AI内容版权与关联原则:在使用AI生成视频或数字人形象时,必须遵循“谁的形象用谁的账号关联”原则。在IP化运营中,确保AI生成的视觉内容与品牌/个人身份的强一致性,避免因形象漂移导致的粉丝信任度下降。", "source_type": "course", "source_label": "小爷账号-2026私域星聚课(配图与课件)", "chapter": null, "section": "课程配图内容.md", "blocks": ["12-3", "07-2"], "skills": ["xy-ai-workflow", "xy-ip", "xy-mode"], "related": null, "q": 2}
{"id": "XY-CF-2362", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "AI驱动的研发加速法:利用生成式AI(Generative AI)与机器学习技术,将传统长周期的研发流程进行“降维打击”。案例显示,原本需要5-8年的新药分子筛选过程,通过AI技术可缩短至18个月。在生意场景中,这意味着对于高壁垒、长周期的行业(如生物科技、新材料),通过引入AI工具可以极速缩短产品从研发到临床/市场化的验证周期,实现“快节奏迭代”对“慢节奏传统模式”的替代。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "06-10"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook"], "related": ["XY-CF-2180"], "q": 4}
{"id": "XY-BM-004", "type": "method", "confidence": "high", "topics": ["IP人设", "认知与心态", "内容创作与平台", "商业案例与实战复盘", "合规与风控", "AI与工具"], "knowledge": "判断一个自媒体账号是否有潜力的三个前提:①内容一致——在发第一条内容前就想清楚每条内容做什么、为什么做、给谁看,所有选题围绕同一人群持续叠加认知资产,而不是追热点导致人设精神错乱②自我开源——内容来自自己真实付过学费、验证过的经历而不是看热点转述,前期没案例时可以直接用「复盘自己的成长过程」本身作为案例③敢于承担风险——敢于表达立场和情绪甚至得罪人,因为AI泛滥让「信息价值」越来越贬值,「情绪价值」(认可观众认可的、反对观众反对的)才是当下内容的突破口", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-1", "03-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-content-scan", "xy-echo-test", "xy-human-touch", "xy-ip", "xy-peer-pick", "xy-precedent", "xy-publish-guard"], "related": null, "q": 3}
{"id": "XY-BM-039", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "免费公开自己的商业方法论/工具反而能带来更大变现——案例:某从业者免费公开全部商业方法论和提示词技能,4个月内间接变现七位数以上;逻辑是先用免费优质内容占领用户心智,让用户对比免费版和市面上的付费产品后主动放弃竞品,最终在「要不要付费上我的课」这个二选一里胜出,而不是在「信息是否免费」这个维度竞争", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3", "08-7"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-content-scan", "xy-peer-pick", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-BM-176", "type": "anti-pattern", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "把某个AI落地角色当成一个需要同时懂业务、技术、产品、写代码的「超级英雄」岗位来招是常见误区——这种全才招不到、招到也难以复制第二个,而且知识全长在这个人脑子里,人一走公司就等于没做过这个项目", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["11-6", "12-4"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-BM-243", "type": "method", "confidence": "high", "topics": ["流量获取", "成交与话术", "AI与工具"], "knowledge": "一种低流量但高变现效率的短视频结构分三步:①用「痒点」筛选(第一句话直接说出目标用户当下的具体困惑,把不相关的流量筛掉,避免转化模型跑偏);②用反常识的独特观点吸引(观点要让用户觉得「原来我之前想的都错了」,独特观点可以直接问AI要「有理有据的反常识观点」);③给「爽点」完成转化(提供具体可操作的解决方案和标准作业流程SOP)——当一个有需求的人遇到一个能解决他需求的人,就可以收钱了", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "09-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-content-scan", "xy-echo-test", "xy-opener", "xy-peer-pick"], "related": null, "q": 3}
{"id": "XY-MA-141", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "持续输出原创内容有两个前提:一是日常养成积累知识和观点的习惯,素材来源可以是读书、刷视频,或和AI对话中被触动的点,把这些积累整理归档成结构化的本地笔记库;据经验,积累到约200条观点或语料时,基本就足以支撑围绕任意话题输出言之有物的内容,且这个积累过程会随着熟练度提升而越写越快;二是要对自己每条内容的可传播范围建立客观预期,主动跳出自己的专业视角,判断不同话题各自能触达的人群体量,避免把某条内容流量差的原因简单归咎为'做了原创'。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "06-9"], "skills": ["xy-ai-workflow", "xy-atomize", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-MA-1564", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "文件名是 AI 判断“要不要读这个文件”的第一依据,名字越精确越不会浪费 token 读无关文件。命名公式:[主题关键词]-[状态/版本].md,例如“用户增长策略-草稿.md”“客户访谈记录-2026-03.md”;避免“优化工具.md”“1.md”这类无法判断内容的名字。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2"], "skills": ["xy-ai-workflow"], "related": null, "q": 3}
{"id": "XY-MA-1565", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "一个文件只做一件事:方法论、每篇文案、笔记分别独立成文件,AI 需要某个信息时只读那一个文件即可,不必为找一段话读几千字的大杂烩。判断标准:文件超过 500 行就考虑拆分。同时层级不超过三层,层级太深人找不到、AI 也要花更多步骤定位。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-ip"], "related": null, "q": 3}
{"id": "XY-MA-1611", "type": "method", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "“知识反向工程”:拿到一段有洞察的优质内容,先提取其明确表达的 5-10 条核心观点(只提原文真说了的,不脑补),再从理论基础、学科背景、方法论工具、实践经验、底层思维五个维度推理形成这些观点背后所需的知识体系,每个知识点写清是什么、如何支撑原文哪个观点、优先级(必修/重要/补充),最后转成分阶段学习目录。要求不生造关联、不堆砌术语,只能靠长期实战获得的知识要诚实标注。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-course"], "related": null, "q": 3}
{"id": "XY-MC-0144", "type": "case", "confidence": "high", "topics": ["流量获取", "内容创作与平台", "AI与工具"], "knowledge": "A类选题通过具体身份/公司结合巨额收入、情感禁忌、极端反差或财务自由等元素来制造吸引力。例如:‘在某公司工作3年赚了2000万’或‘被裁员后3个月赚了100万达’。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-1", "07-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 3}
{"id": "XY-BKG-001", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "反常识归因开头:先点破读者以为的失败原因,再否定它、给出真正原因,最后承诺解法清单——这个结构比直接讲道理更抓人,因为先戳中了读者的自我怀疑(我是不是哪里做错了),再给出一个意外的答案。某AI工具教程类爆款视频用「你可能不是不会用,而是根本没打开隐藏能力」开头,播放136万,收藏率3.2%(远高于同类视频,说明观众把开头的承诺当真、真的收藏等着照做);另一条同类视频用「不是模型不行,是任务太乱」开头,播放79万,收藏率2.3%,两条视频用的是同一个「否定表面原因-给出真实原因」结构,交叉印证了这个开头模式的有效性。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 3}
{"id": "XY-BKG-325", "type": "case", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "真实口播稿里会用拟声叠词(「啪啪啪啪啪」「搞搞搞搞搞搞搞搞」)直接模拟动作发生的节奏和语速,这是书面语/AI生成文案几乎不会自然产出的表达方式(AI倾向于用「接着」「随后」这类规范连接词)。同批里出现这类拟声叠词的多条视频播放量并不低,有的达到3.66万,说明这种「不像写出来的」表达并不影响传播效果,反而可能是真实感的来源之一。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-7", "06-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 3}
{"id": "XY-BKG-326", "type": "case", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "真人口播稿会高频插入「理解吗?」「对吧?」「你懂吗?」这类反问语气词,不是为了真的等待回答,而是模拟面对面讲话时确认对方跟上没有的习惯性停顿;AI生成文案通常会省略这些「无信息量」的口头禅,导致文字读起来「太顺」反而不像真人在说话。含大量此类反问的教程视频播放量达13.1万。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 3}
{"id": "XY-BKG-607", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "开头用「这是被同行/圈内人瞧不起的生意」这种身份反差感切入,把一个看起来低端的项目包装成我本来不屑做但发现门槛低到离谱的反常识发现,比直接介绍项目本身更能激发好奇心和优越感;某讲AI智能体项目的视频用这个开头,播放184万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-1", "13-1"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 3}
{"id": "XY-BKG-612", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "长时长讲解视频用第一个案例、第二个案例、第三个案例或三个亮点/四个观点这种数字化清单结构组织全篇,观众即便中途划走一段,也能靠序数词迅速定位自己听到哪了、还剩多少——这是长视频(十几分钟以上)防止中途流失的核心节奏工具;多条AI工具测评类爆款视频(其中一条播放740万)都采用这个结构", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "12-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-script-glue"], "related": null, "q": 3}
{"id": "XY-BKG-616", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "结尾用开放式提问(你觉得你的护城河是什么)代替喊话式call to action(扣评论、点关注),开放式提问把决定权交给观众,反而更容易激发真实评论而不是无意义的口号刷屏;某讨论AI对就业冲击的视频用这个收尾,播放17万、评论率明显高于同类", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["09-4", "13-2"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-BKG-620", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "讲完一个操作环节后,立刻用「我们来看一下最终效果」做验收,再进入下一个环节,形成预告—执行—验收的固定节奏循环;这个循环重复三次左右(比如演示三个案例)比一次性讲完所有操作再统一展示效果更抗中途流失,因为观众每隔一小段就能获得一次完成感的反馈;某AI建站工具测评视频用这个结构反复三次,播放740万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "12-2"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 3}
{"id": "XY-BKG-622", "type": "insight", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "外行做得比专家更好这个反常识现象,能精准安慰「我不懂技术是不是就用不好AI」这批人的焦虑——当观众看到一个完全没有技术背景的人反而做得比程序员更好时,会产生技术门槛没那么重要的解脱感,这种解脱感是转发和收藏的主要驱动力,而不是单纯的知识增量;某AI协作方法论视频用了这个反差案例,播放275万、收藏17万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["13-1", "12-1"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-echo-test"], "related": null, "q": 3}
{"id": "XY-BKG-625", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "用排比式的具体场景反问句(写短视频脚本的做一个短视频智能体、写朋友圈的做一个朋友圈智能体……)让不同职业背景的观众都能在其中找到自己对应的那一句,排比结构本身就是在批量制造这说的就是我的共鸣点,命中率比单一案例更高;某企业AI落地科普视频用这个结构,播放38万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "07-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-BKG-640", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "解释一个技术流程时,数字表达上保留口语化的不精确(200多刀、七八百、一两百块这种带模糊区间的说法),而不是精确到个位数的书面语数字——这种模糊区间感更接近真人凭记忆口述,AI生成的文本反而倾向于给出看似精确实则编造的具体数字", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-3", "06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 3}
{"id": "XY-BKG-933", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "过去做展示专业的内容习惯把营销目的藏起来(讲案例、放客户证言),但作者发现在内容里主动承认「我讲这段某种程度上也是为了证明自己」、把思考过程和判断逻辑完整摊给观众看,观众反而会觉得「很真诚」而不是「营销号」。真人感有一部分来自愿意暴露自己的动机,而不是伪装成纯粹利他。这条视频播放99399、分享1287。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-2", "06-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 3}
{"id": "XY-BKG-1573", "type": "method", "confidence": "high", "topics": ["流量获取", "内容创作与平台", "AI与工具"], "knowledge": "小红书推流机制是阶梯式测试:笔记发布先进审核,通过后系统AI识别内容打标签给200-500曝光做小流量测试,数据好进下一流量池,24小时内数据不好或同质化内容超过70%就直接限流。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["05-1", "06-8"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-private-ops", "xy-traffic"], "related": null, "q": 3}
{"id": "XY-MB-477", "type": "principle", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "\"不是X,而是Y\"是典型的\"AI味\"句式,但这个句式本身不等于烂文案。它变差的原因通常是:X应该是戳中观众原有认知的普遍误解,当观众本就认同X是对的时候,作者却断言\"不是X\",就会让文案显得无的放矢、缺乏说服力。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 3}
{"id": "XY-CE-1415", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "数据价值链分层布局法:将大数据产业分为“数据生成(传统IT)、数据整合(系统集成)、数据应用(AI/分析)”三个层级。在生意场景中,初创企业应根据技术难度与市场成熟度选择切入点:在数据源多元且质量参差的“整合层”寻找机会,或在应用场景分散、格局未成的“应用层”利用细分领域数据优势进行突围。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "13-3"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 3}
{"id": "XY-CE-1949", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "AI驱动的业务增长路径:营收提升与成本降低的双轮驱动。根据行业调研,应用AI可以从两个维度直接影响利润表:一是通过精准营销和产品开发提升营收(平均67%的企业受益),二是通过优化供应链、服务运营及人力资源流程降低成本(平均79%的企业受益)。在私域生意中,这意味着AI的应用不应仅停留在“写文案”的工具层面,而应深入到“用户生命周期价值(LTV)预测”和“自动化库存/物流管理”这两个核心利润点中。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "09-5"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CE-1957", "type": "method", "confidence": "medium", "topics": ["私域运营", "认知与心态", "AI与工具"], "knowledge": "可解释性分层模型(Explainability Layers):在应用AI或复杂决策系统时,需从三个维度构建信任。1.工程师可解释性:解决技术实现逻辑;2.因果可解释性:解释输入与输出的关联(为什么这么做);3.信任诱导可解释性:提供用户感知层面的信息以建立信任。在私域运营或高客单价生意中,当算法驱动推荐或定价时,必须具备“信任诱导可解释性”,即告知用户“为什么为你推荐”或“为何此价格合理”,否则复杂的技术黑盒会直接摧毁用户信任。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["07-2", "08-7"], "skills": ["xy-ai-workflow", "xy-playbook", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-CE-1992", "type": "case", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "工业化驱动的业务增量案例。通过将机器学习技术引入核心业务流程,不同行业展现了显著的规模化价值:某全球制药公司通过集成个性化推荐系统,在超过50个国家和药物组合中实现了10-20%的销售增长;某亚洲数字银行通过工业化AI用例,实现了每年超过150个业务场景的落地;某北美零售商通过自动化部署流程,将模型上线时间从数月缩短至分钟级。这证明了当技术实现“工业化”后,其核心价值在于通过极高的资产复用率和部署效率,实现业务规模的指数级扩张。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["13-1", "11-12"], "skills": ["xy-ai-workflow", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-CE-2003", "type": "case", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "自动化测试与部署的效能飞跃:通过引入先进的DevOps工具和基础设施即代码(IaC),可以实现业务逻辑的极速交付。案例显示:某大型零售商通过基础设施即代码,将部署时间从数周缩短至30分钟;某金融机构利用AI工具生成单元测试,使核心后端应用的测试编写速度提升了180倍。这启示生意操盘手,在规模化扩张阶段,必须通过“标准化组件”和“自动化流程”来解决重复性劳动,避免人力成本随业务复杂度呈指数级增长。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "11-8"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-CE-2013", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "技术泡沫陷阱(Supremacy Claims Trap):警惕过度夸大技术能力的“性能幻觉”。目前许多声称超越传统算力的案例在实际业务场景中可能并不成立,因为未计入额外的纠错成本或运维开销。在商业决策中,如果一个新工具(如某种新型流量算法、AI模型)宣称能颠覆现有逻辑,但未在特定利基场景(Niche Tasks)中通过实际业务指标验证,则需警惕其可能带来的高昂试错成本。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-4"], "skills": ["xy-ai-workflow", "xy-question-spec"], "related": null, "q": 3}
{"id": "XY-CE-2074", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "技术泡沫陷阱:在行业爆发初期,容易出现“全线押注”的错误。历史经验显示(如2006-11年的清洁能源泡沫),并非所有新技术都能存活,部分技术会因为执行风险过高或成本无法覆盖而退出。对于生意人而言,在面对新兴趋势(如AI、新能源)时,应警惕“全盘接受”的盲目感,要识别哪些是具备确定性成本优势的“硬通货”,哪些是仍处于高风险实验阶段的“伪需求”。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-4", "13-2"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 3}
{"id": "XY-CE-2132", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "AI技术应用场景的分类逻辑:在构建数字化生意时,应根据技术特性匹配业务需求。主要包含四类核心能力:1. 机器学习(ML):通过统计学从历史数据中学习,适用于排班优化、库存预测等逻辑化场景;2. 计算机视觉(CV):处理图像/视频信息,适用于人脸识别、自动化质检等感知场景;3. 自然语言处理(NLP):理解和生成文本/语音,适用于智能客服、文案自动生成等沟通场景;4. 深度强化学习:在不确定环境中做决策,适用于复杂路径规划或自动化控制。操盘手应根据业务环节是需要“感知”、“理解”还是“决策”,来选择对应的技术工具。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "06-11"], "skills": ["xy-ai-workflow", "xy-playbook"], "related": null, "q": 3}
{"id": "XY-CE-2208", "type": "insight", "confidence": "medium", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "技术迁移的兼容性陷阱(Legacy System Compatibility)。当企业试图通过引入AI模型或新型数字身份技术来升级私域链路时,必须预判“碎片化解决方案”带来的整合难度。在实际生意场景中,新技术的落地往往受限于旧有的业务流程、分散的第三方合作伙伴以及缺乏统一标准的现有系统。操盘者在规划数字化升级路径时,不能仅关注新技术的ROI,必须将“技术兼容性”和“组织规范转型”纳入成本评估。如果新旧系统无法实现数据互通,所谓的数字化升级只会增加运营的复杂度和隐性成本。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "11-1"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CE-4720", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "全链路数据集成法:在复杂业务体系中,通过ESB(企业服务总线)或EAI(企业应用集成)技术,将生产、营销、财务、项目管理等多个孤岛系统通过统一协议(如Webservice)进行连接。在生意场景中,这意味着当你的私域流量产生订单时,后端库存、财务账务、物流配送能实时同步,避免人工对账和信息差导致的交付失败。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "12-2"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CE-5399", "type": "principle", "confidence": "medium", "topics": ["私域运营", "团队与模式设计", "AI与工具"], "knowledge": "智能化降本增效原则:利用自动化技术(如RPA或AI)处理非结构化数据与重复性劳动,核心逻辑在于释放高价值人力。在实际生意场景中,通过自动化工具处理后台资料录入、数据核对及客户信息搜集,可实现如“1年内扩展10倍规模同时降低40%成本”的杠杆效应,并将人工错误率从1%-1.5%降低至小于1%,从而实现前中后台的直通式处理。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "11-6"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-CE-5437", "type": "method", "confidence": "medium", "topics": ["私域运营", "团队与模式设计", "AI与工具"], "knowledge": "RPA自动化分阶法:将业务流程分为“现阶段”与“目标阶段”。第一阶段通过RPA(机器人流程自动化)解决重复性数据录入、初步核实与标准化作业,实现24/7无人干预;第二阶段引入机器学习(ML)实现自动数据提取、文档分类与合规规则校验。在生意场景中,当业务规模扩大导致人工处理成本激增时,应先用RPA替代80%的重复录入工作,再通过AI提升复杂文档的处理能力。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "10-2"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-CE-5903", "type": "anti-pattern", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "自动化误区:过度依赖工具实现业务目标。在某些BPO(业务流程外包)或运营场景中,客户的核心痛点往往不在于“自动化工具”本身,而在于“流程与运营逻辑的定义”。如果业务底层逻辑(Process & Operations)是混乱的,单纯引入自动化技术不仅无法解决问题,反而会放大错误。生意场景判断:在搭建私域运营体系或交付标准化服务时,应先完成业务流程的标准化(SOP)与逻辑闭书,而非在流程尚未理顺时就急于引入复杂的自动化/AI工具。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-11", "06-10"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-CE-5948", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "大客户分层与定制化方案法(Account Planning & Tailored Solutions):在医疗器械或高客单价B2B生意中,不应试图覆盖所有客户,而应识别出贡献了绝大部分市场潜力的核心账户(如案例中前3大客户占据75%的潜力)。针对这类头部客户,应提供定制化解决方案(如共同开发、共同制造或成本分摊模式),通过深度绑定核心客户来快速切入高壁垒市场。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["09-1", "11-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CE-5967", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "场景化成本/效率模型(Co-development vs In-house):在产品开发决策中,需权衡“自主研发”与“联合开发(Co-development)”的成本与速度。案例显示,某药企自主研发注射笔耗资约500万人民币,而选择与专业厂商合作则需投入约1000万人民币。对于中小规模的生意操盘手,必须意识到:选择合作虽然短期成本翻倍,但能快速获得成熟的供应链能力和技术标准;而自主研发虽然省钱,但可能面临开发周期长、技术门槛高导致的市场机会流失。决策逻辑应从“单纯看研发成本”转向“投入产出比与市场进入速度的平衡”。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-4", "11-1"], "skills": ["xy-ai-workflow", "xy-selection"], "related": null, "q": 3}
{"id": "XY-CE-6492", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "PEST宏观环境分析:从政治(P)、经济(E)、社会(S)、技术(T)四个维度审视外部环境对生意的长期影响。在私域或品牌经营中,需关注政策法规(如数据合规)、消费趋势(如老龄化或情绪价值需求)及技术变革(如AI工具应用),判断未来3-5年行业是红利期还是收缩期,从而决定战略投入的规模。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-4"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 3}
{"id": "XY-CF-1707", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "业绩管理五步法:通过建立清晰指标、目标分解设定、有效跟踪业绩、进行积极的业绩对话以及问题解决,形成闭环管理。在生意场景中,这要求管理者不仅要看最终利润结果,更要将目标拆解至班组/岗位层级的行为指标(KAI),并通过看板实现目视化,确保一线员工知道“做什么”以及“做对没”。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-7", "11-10"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-playbook"], "related": null, "q": 3}
{"id": "XY-CF-2076", "type": "case", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "能源巨头效率优化案例:某大型电力生产商通过构建多层神经网络模型,利用过去两年的工厂运行数据进行训练,旨在寻找不同环境因素(如温度、湿度)下的最优控制参数。通过将AI模型验证结果转化为每30分钟一次的实时操作建议,成功实现了燃烧器使用量降低30%,并最终在全公司范围内节省了2300万美元。这证明了“数据驱动决策+实时反馈闭环”在重资产或高频业务场景中具有极高的ROI。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["13-1", "12-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-CF-2164", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "机器人/自动化工具的效能评估模型:衡量一个新工具(如AI、自动化设备或新SOP)是否具备大规模替代人工的潜力,需从三个维度进行判断:1. 灵活性(Dexterity),即工具处理复杂、非标任务的能力是否接近人工;2. 续航/持续性(Battery Life),即工具在不进行维护或干预的情况下,能否覆盖完整的业务工作周期(如8小时);3. 自主决策力(Autonomy),即在遇到异常情况时,工具能否通过内置逻辑进行自我纠错或决策而非必须人工介入。在生意场景中,老板判断是否引入新工具时,不能只看单次效率提升,必须评估其在“非标场景”下的容错率和全天候作业能力。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-11", "06-10"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CF-2179", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "基于多维指标的技术趋势评估模型。判断一个新兴技术(如AI+生物工程)是否具备商业爆发潜力,应结合四个维度进行交叉验证:1. 资本流向(Equity investment);2. 技术产出(Patents/Research);3. 市场搜索热度(Searches);4. 人才需求强度(Job postings)。当这四个指标出现分化时,例如“专利数与搜索量上升”但“投资额下降”,往往预示着该领域正处于从“资本驱动”转向“技术/应用驱动”的转型期,是寻找高壁垒细分赛道的关键窗口。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-4", "13-3"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 3}
{"id": "XY-CF-2180", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "AI驱动的研发加速法:利用生成式AI与机器学习技术,将传统长周期的研发流程(如新药分子筛选)从5-8年缩短至18个月。在生意场景中,这代表了“技术杠杆”对生产力的重塑,当你的核心产品研发周期过长导致现金流压力大或错过市场窗口时,应评估是否可以通过引入成熟的AI模型进行“降维打击”,将研发成本从重资产投入转向算法驱动。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "06-10"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": ["XY-CF-2362"], "q": 3}
{"id": "XY-CF-2209", "type": "principle", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "技术成熟度与市场准入的平衡原则:新兴技术的商业化落地受限于成本、基础设施和人才供给三个维度。材料指出,尽管新能源技术前景广阔,但规模化过程中的高成本、电网基础设施建设滞后以及技能型劳动力(Skilled Labor)的短缺是核心障碍。在进行生意布局时,不能仅看技术趋势(Trend),必须同时评估该技术的“配套能力”——即你的供应链、交付团队以及基础设施是否能支撑起技术带来的规模化需求,否则极易陷入“有技术、无市场”的资金空转陷阱。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8", "10-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CF-2244", "type": "case", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "AI驱动的客户服务效能升级案例:某全球金融机构通过引入定制化生成式AI聊天机器人,成功解决了传统客服工具的瓶颈。在应用初期,该方案使20%的用户避免了长时间等待,并实现了即时反馈。对于私域生意而言,这证明了AI在处理高频、标准化的客户咨询(如物流查询、基础售后)时,能显著提升用户体验并释放人工客服压力。通过将AI从“简单问答”升级为“个性化辅助”,可以实现服务规模的指数级扩张。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "13-1"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-CF-2258", "type": "case", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "能源巨头的AI增效案例:某大型能源公司通过构建AI中心,每日处理超过50亿个数据点,利用实时预警和预测模型优化钻井路径与设备维护。通过引入名为“热效率优化器(HRO)”的解决方案,该公司在验证模型准确性后将其推广至全线业务,最终实现了26个厂区、67台机组的规模化应用,直接节省了约2300万美元成本。这证明了“模型验证+规模化复制”是AI从实验走向大规模盈利的必经之路。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["13-1", "11-12"], "skills": ["xy-ai-workflow", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-CF-2259", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "AI落地三步走路径:1. 解决数据组织问题,确保数据的可用性与治理水平;2. 实现技术集成,利用即插即用的解决方案实现标准化与互操作性;3. 强化变革管理,通过持续学习、最佳实践传播和角色示范来驱动组织层面的适应。对于生意操盘手而言,这意味着在引入新技术时,必须同步进行“数据资产化”与“员工技能升级”,否则技术投入将无法转化为实际的ROI。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8", "11-6"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CF-2270", "type": "anti-pattern", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "技术落地中的“度量缺失”陷阱:在引入新技术(如AI工具)时,如果缺乏清晰的开发者生产力衡量指标、整合挑战以及大规模的人才重塑计划,技术升级往往会演变为成本黑洞。对于生意操盘手而言,引入任何新系统或新流程前,必须先定义好“效率提升”的具体量化指标(如转化率、响应时长),否则技术投入将无法转化为实际利润。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-12", "11-10"], "skills": ["xy-ai-workflow", "xy-close", "xy-goal-card", "xy-mode", "xy-ops"], "related": ["XY-BK3-1461"], "q": 3}
{"id": "XY-CF-2325", "type": "anti-pattern", "confidence": "medium", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "“试点陷阱”与规模化鸿沟:许多企业在引入新技术(如边缘计算或AI模型)时,容易陷入“能跑通Demo但无法规模化”的困境。这种现象通常由三个因素驱动:一是缺乏即插即用的标准化解决方案,导致每个新场景都要重复建设;二是技术栈与现有业务系统的集成成本极高;三是缺乏对长周期回报的预期管理。对于生意操盘手而言,如果一个技术方案在从“试点(Pilot)”转向“规模化落地(At-scale)”时,需要投入不成比例的定制化开发成本和人力资源,那么该技术在当前阶段可能并不具备商业扩张性。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8", "01-3"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 3}
{"id": "XY-CS-3970", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "数字化转型中的“效率杠杆”应用:在传统制造业或重工业领域,通过引入AI和自动化技术(如智能制造、预测性维护),可以将生产效率提升1.0%-1.5%甚至更高。在利润极其微薄的存量市场,这种“极小值”的效率提升往往就是决定生死的分水岭。操盘手应寻找业务中“重复性高、容错率低”的环节进行技术改造。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "11-6"], "skills": ["xy-ai-workflow", "xy-ops", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-TF-001", "type": "method", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "千川素材追投的筛选与分组打法:在当日整体消耗前 100 的素材里,按「CTR>3% 且 ROI>2」勾选合格素材,约 10 个一组建追投计划;分组差异化配置——头部组预算 200 元/调控 5 小时,次组 200 元/3 小时,尾组 100 元/3 小时。符合条件的素材不足时如实报缺口,不硬凑。具体阈值是参考参数,各品类需按自己数据实测校准。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-3"], "skills": ["xy-ai-workflow", "xy-traffic"], "related": ["XY-TF2-0061"], "q": 3}
{"id": "XY-TF-006", "type": "method", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "控成本计划的自动 ROI 调节节奏示例:半小时消耗不满 100 元说明跑不动,目标 ROI 下调 1 放量;半小时消耗超 300 元说明跑太猛,目标 ROI 上调 1 收口;每日预算消耗低于 80% 时自动补预算。本质是把「消耗速度」当作出价合理性的实时反馈信号,用小步幅高频次代替人工大改。参数按自己盘子校准。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-5", "04-4"], "skills": ["xy-ai-workflow", "xy-traffic"], "related": null, "q": 3}
{"id": "XY-IPB-0394", "type": "case", "confidence": "medium", "topics": ["流量获取", "商业案例与实战复盘", "AI与工具"], "knowledge": "据案例描述,在电商AI内容生产场景下,AI工具的商业价值在于其成本优势与规模化能力。如果AI能将人力成本降低至原有的1/5(例如从200元降至20-30元),且能通过大规模、高频次的自动化生产(如每天发数万条视频)来获取流量和GMV,则具备极强的商业变现能力。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["13-1", "06-11"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 3}
{"id": "XY-TF2-0033", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "直播全域投放五维核心指标监控法:在进行本地推(抖音本地生活/同城业务)投放时,必须通过五个核心维度来判断生意健康度:1.消耗(投入成本);2.全域成交订单数(规模指标);3.全域成交金额(GMV规模);4.全域支付ROI(投入产出比,核心效率指标);5.全域成交订单成本(单客获取成本)。在实际操盘中,严禁通过其他指标反推计算ROI或订单成本,必须以平台后台实时显示的原始数值为准。判断逻辑在于:当消耗增加时,需同步观察ROI与订单成本的变化趋势,若ROI下降且订单成本上升,说明流量效率在降低。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-6"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-TF2-0034", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "数据稳定性校验法(Three Stable Card Snapshots):在进行自动化数据采集或业务复盘时,针对动态变化的指标(如直播间实时消耗、ROI等),不能仅凭单次抓取结果下结论。必须要求数据在连续三次读取中保持稳定(Stable),即排除掉页面加载延迟、DOM结构变动或数值跳动带来的干扰。在生意场景中,这意味着当你在观察直播间实时成交额或ROI时,必须确认数据已“settle”(稳定),避免因网络波动或系统延迟导致的错误决策,确保后续所有策略动作基于真实、稳定的数据基准。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["05-6", "04-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops", "xy-private-ops", "xy-question-spec"], "related": null, "q": 3}
{"id": "XY-TF2-0044", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "直播全域调控三阶法:通过“一键起量”、“精细调控”与“搜索抢首屏”三种逻辑实现流量覆盖。一键起量侧重于短周期(0.5-24小时)的爆发式预算投放;精细调控侧重于以转化成本(CPA)或投产比(ROI)为核心目标的稳定增量;搜索抢首屏则通过手动或自动竞价结合关键词拦截,实现精准流量截流。在实际生意操盘中,应根据直播间当前阶段(起号期、爆发期、平稳期)切换对应的调控工具,而非单一依赖一种模式。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["05-4", "03-6"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-TF2-0050", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "参数校验与异常捕获机制(Guard Mechanism):在进行高频、关键的业务数据修改(如调整广告日预算或ROI目标)时,必须在执行动作前后建立“状态校验点”。具体流程为:1. 触发编辑指令;2. 等待并确认输入框可见(Visible);3. 执行修改动作;4. 再次校验界面是否出现“确认/保存”按钮。在生意实战中,这意味着当你通过自动化工具或人工批量调整投放策略时,不能仅执行“点击”动作,必须建立一套反馈闭环,确保每一个预算指令都真正进入了系统的待提交状态,否则会导致策略执行失效或数据错位。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-5", "04-4"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 3}
{"id": "XY-TF2-0051", "type": "principle", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "数值一致性校验原则(Numeric Consistency Principle):在处理涉及金钱、比例等核心经营指标的输入时,必须执行“输入-读取-对比”的三步校验法。即:执行填入动作后,必须重新读取输入框内的数值,并将其与目标值进行标准化(Normalize)对比。如果两者不一致,说明输入失败或被系统拦截。在生意场景中,这适用于对投放ROI、单客成本(CAC)等核心指标的批量调优,防止因系统延迟或输入框焦点丢失导致的错误指令执行。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-4", "12-3"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 3}
{"id": "XY-BM-001", "type": "method", "confidence": "high", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "AI agent团队管理的三个核心能力:①主动性(通过心跳机制定时自查进度和卡点主动汇报,而不是等指令才动)②可重入性(网络中断/长任务打断后能从中断点稳定恢复,甚至能自我修复导致中断的原因)③「AI式思考」(面对同一任务优先选择最高效的API调用方式而不是模拟人类操作,比如发推文该调用平台API而不是模拟浏览器点击输入)", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-peer-pick", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-BM-078", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "用AI Agent搭建4层信息结构来跟踪个人重大决策:①原始信息(如实记录当事人的原话,标注可信度,不代表客观事实)②决策事实(明确已发生的决定,视为确凿)③假设推断(分别记录当事人自己的假设和Agent的假设,标注为待验证,不预设对错)④结果验证(事后回填现实结果,提炼核心教训、暴露出的个人行为模式和未来应对规则)", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-2", "12-4"], "skills": ["xy-ai-workflow", "xy-casefile", "xy-peer-pick"], "related": null, "q": 2}
{"id": "XY-BM-145", "type": "method", "confidence": "high", "topics": ["选品逻辑", "IP人设", "AI与工具"], "knowledge": "把踩过的坑和调好的流程沉淀成可复用的AI技能(skill),验证有效性要做到三点:①用真实任务做对比测试(带skill和不带skill各跑一遍打分对比,而不是主观觉得写清楚了)②测试者必须是不带任何记忆的干净智能体,自己看得懂不算数③反复打磨触发描述,因为这是技能能否被正确唤醒的唯一开关", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-ip", "xy-peer-pick", "xy-selection"], "related": null, "q": 2}
{"id": "XY-BM-146", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "在动手做AI自动化开发之前,先做头脑风暴和POC验证来消除方向上的模糊性——AI和传统敏捷开发不同,它容易第一步就直接给出一个方向可能错误的完整方案,不像敏捷那样每一步都能验证调整;慢一步把方向想清楚,能省下后面反复返工的成本", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-2", "06-10"], "skills": ["xy-ai-workflow", "xy-peer-pick"], "related": null, "q": 2}
{"id": "XY-BM-155", "type": "principle", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "真正能自己进步的Agent系统需要建立记录-评估-优化的自迭代闭环——把每次执行过程中哪一步做对哪一步做错、人工修正后的正确做法都沉淀下来作为下一轮优化的训练数据,否则Agent交付时是什么水平,半年后可能还是这个水平,甚至因模型和外部环境变化而表现漂移", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-4", "12-2"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-BM-173", "type": "anti-pattern", "confidence": "high", "topics": ["内容创作与平台", "商业案例与实战复盘", "AI与工具"], "knowledge": "企业AI项目常见的三种死法:①数据没打通(demo用的是手工清洗过的干净样例数据,真实业务数据分散在多个系统甚至员工个人电脑里)②没有能算清ROI的场景(选题定得又大又虚,业务方无法判断是否该持续投入)③没有人对上线后的结果负责(交付完就走,出问题没人兜底改进)", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-1", "11-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-peer-pick", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-BM-174", "type": "method", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "可靠的AI落地路径是四步:先选一个窄场景切入而不是一上来做大平台;只打通这个场景需要的数据而不追求一次治理全公司数据;派人常驻业务里对结果负责而不是交个demo就撤;一个场景跑通算清账后再复制到下一个场景", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-1", "11-8"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-BM-181", "type": "method", "confidence": "high", "topics": ["IP人设", "AI与工具"], "knowledge": "企业一号位不需要亲自搭建Agent或参与具体培训执行,但必须牢牢抓住AI转型的所有权——定基于业务增长的结果目标、给转型负责人真正的资源/规则/否决权(避免各部门表面支持实际互相拖延的「假授权」)、在组织调整和抽调核心业务人员时亲自做出关键取舍", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["11-10", "11-7"], "skills": ["xy-ai-workflow", "xy-ip", "xy-peer-pick"], "related": null, "q": 2}
{"id": "XY-BM-242", "type": "method", "confidence": "high", "topics": ["流量获取", "选品逻辑", "内容创作与平台", "AI与工具"], "knowledge": "长期稳定做原创要建立「由内而外」的选题习惯——基于自己的产品和用户调研,推导出几十条到上百条能获客的选题库;为了保证这些选题也有流量确定性,可以定期收集「亲和性赛道」(目标用户同时也会关注的相邻领域,比如做自媒体培训的用户往往也关注商业思维、AI科技类内容)里表现好的爆款开头(小博主5万赞以上、大博主20万赞以上为标准),把这些开头结构模板化后套进自己的业务内容里——因为跨赛道不构成抄袭,又能持续保证内容开头的吸引力", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-1", "07-5"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener", "xy-peer-pick", "xy-selection"], "related": null, "q": 2}
{"id": "XY-MA-1177", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "高效原创的大前提是养成日常储备知识和观点的习惯:观点来源可以是读书、刷视频、和AI聊天被触动的内容;用手机备忘录或flomo这类随手记工具记录,再导出为markdown文件(或直接用Obsidian)打包进专门文件夹,形成个人本地知识库。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MA-1185", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "原创方法二:借用跨赛道爆款的开头(甚至第一句话)直接输出自己业务范畴的认知。案例:某美业赛道博主的爆款讲“审美不好的人在提供审美服务”,作者沿用同样的开头结构讲自媒体领域“审美好的人执行力普遍有问题”,再用AI结合自己沉淀的“创作欲望大于创作能力”等观点迅速产稿拍摄。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener", "xy-playbook"], "related": null, "q": 2}
{"id": "XY-MA-1335", "type": "method", "confidence": "high", "topics": ["IP人设", "内容创作与平台", "AI与工具"], "knowledge": "阶段七“信任积累”:用户已认定需要专业帮助,但在观察“你靠不靠谱、找你有没有风险”。内容任务是消除最后一道风险顾虑,完成从“想找”到“找你”的跨越,选题如揭穿AI营销留学文案、揭秘国际本科连环诈骗、真实学员逆袭案例复盘(“GPA不高凭啥逆袭名校,走对了哪几步”)。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["13-2", "08-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-playbook"], "related": null, "q": 2}
{"id": "XY-MA-1486", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "整理对标文案的第一步只做一件事:把原始内容整理干净,保留原意、不改写、不扩写,先不让AI分析也不让它仿写。拿到干净文本之后,拆结构、仿写、提炼选题等分析动作再另行进行——先分离\"采集\"与\"分析\"两个环节。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-10", "06-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-peer-pick"], "related": null, "q": 2}
{"id": "XY-MA-1511", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "降低手抄式内容制作成本的做法是把流程拆成两段:先把口播稿喂给 AI(豆包、DeepSeek、Claude、GPT 均可)生成板书排版方案,再把实际手写环节外包给二手交易平台上的代抄/大学生手抄服务,自己只保留选题和稿件。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-11", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-peer-pick", "xy-traffic"], "related": null, "q": 2}
{"id": "XY-MA-1520", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "微信到 Obsidian 的同步工作流:在 Obsidian 装一个第三方同步插件并开启定时自动同步,把插件配套的微信个人号加为好友,之后随手把文字、图片、文章转发给它,内容就会自动落到本地笔记库;这样微信成为随手采集入口,Obsidian 成为集中沉淀的仓库。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "08-9"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-MA-1553", "type": "definition", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "“圆桌思想家”是一种深度分析框架:模拟苏格拉底、亚里士多德、费曼、德鲁克四位思想家的圆桌研讨,对一个议题做五阶段结构化分析。目标不是给出单一答案,而是构建深度理解网络、揭示核心矛盾与探索方向。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-6"], "skills": ["xy-ai-workflow", "xy-roundtable", "xy-term-crack"], "related": null, "q": 2}
{"id": "XY-MA-1572", "type": "method", "confidence": "high", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "复合型业务的文件系统采用三层结构:想法工作台(所有想法的源头和协作深化区)、通用知识库(跨业务的方法论和素材库)、业务板块(各业务独立的生产线)。想法在工作台深化后分发到各业务板块独立发展。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "11-11"], "skills": ["xy-ai-workflow", "xy-idea-desk", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-MA-1592", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "AI 整理深化稿素材时要完整读完对话记录,按板块提取时保留对话中的具体措辞、案例、类比、交锋,不压缩成概括性语言;质量标准是写终稿的人拿到深化稿不需要再翻对话记录就能直接开工。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "06-10"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MA-1612", "type": "method", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "陪学时根据学员在检查站的回答自适应推进:完全掌握(准确且能举一反三)→下一课;基本掌握(大致正确有小偏差)→指出偏差在哪、为什么偏,再下一课;部分理解(概念混淆)→用全新角度和例子重讲,另开补充课;尚未理解→拆到更基础层面,从学员已经懂的地方重新搭起。每课末尾设检查站(概念确认题、新场景应用题、思辨题),并维护进度文件记录每课掌握情况和关键误区。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-7"], "skills": ["xy-ai-workflow", "xy-course"], "related": null, "q": 2}
{"id": "XY-MC-0063", "type": "method", "confidence": "high", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "在实际应用中,应优先将核心规则和指令写进系统提示词;只有当内容无法容纳时,才考虑使用知识库。对于Agent类应用,可尝试用原始文档代替RAG知识库并建立数据权威索引。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "12-2"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-MC-0067", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "采用“广度调研(Wide Research)+ 深度研究(Deep Research)”的三步流程进行素材收集:首先通过广度调研获取大量结构化素材,接着让AI从结果中挑选关键点,最后结合挑选出的关键点与原始素材进行深度研究,实现从海量信息到精炼深度内容的转化。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "13-1"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0075", "type": "method", "confidence": "high", "topics": ["IP人设", "内容创作与平台", "AI与工具"], "knowledge": "在进行话题拆解时,应先分析话题属性再选择专家角色。例如面对“网红与IP的区别”这类题目,应从粉丝关系质量、竞争结构可持续性、内容资产化三个视角切入。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-6", "13-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip"], "related": null, "q": 2}
{"id": "XY-MC-0082", "type": "principle", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "掌握学习的核心在于诊断性反馈:诊断的目的不是为了打分,而是为了精准定位学习者在哪个细微环节卡住了。只有通过细颗粒度的补救,才能避免在知识地基不牢的情况下盲目推进。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-7"], "skills": ["xy-ai-workflow", "xy-ip"], "related": null, "q": 2}
{"id": "XY-MC-0084", "type": "method", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "交互式学习适用于听过方法但不会操作、需要系统化消化课题、以及需要持续推进以防止遗忘的学习场景。其核心在于将知识从“听过”转化为“会做”。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-7", "12-4"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-MC-0119", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "一种有效的短视频结构是:通过断言对立引发冲突,接着进行解释,随后引入客户案例,最后以金句小结并进行反问自检。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-MC-0120", "type": "method", "confidence": "high", "topics": ["IP人设", "内容创作与平台", "AI与工具"], "knowledge": "对话推演型结构是一种替代传统教学式的结构,通过直接拆解错误案例并现场演示修改过程,以“共同推演”而非“单向教学”的姿态引导观众,通过主线逻辑层层深入。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-opener"], "related": null, "q": 2}
{"id": "XY-MC-0122", "type": "anti-pattern", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "传统的‘先讲理论再举例’的教学式结构容易显得说教,且容易稀释核心观点。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": ["XY-ZB-3010"], "q": 2}
{"id": "XY-MC-0123", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "对话推演型结构通过‘拿一段烂口播,四步改造,改完即金句’的逻辑,将教学过程转化为共同推演的过程,增强互动感与逻辑深度。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0124", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "一种高效的文案改稿逻辑:第一步将抽象理论转化为形象的比喻;第二步将干巴巴的观点转化为带有情绪色彩的表达。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0125", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "一种常见的文案创作套路是:通过断言对立引发关注,随后进行解释,结合客户案例,最后通过金句总结并进行反问自检。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0126", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "通过四个维度的转换实现文案改稿:将理论转化为比喻,将观点转化为情绪,将思考转化为故事,将方法转化为可验证的画面。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0147", "type": "principle", "confidence": "high", "topics": ["流量获取", "内容创作与平台", "AI与工具"], "knowledge": "开头必须在极短时间内(2-10秒)抓住注意力,且开头承诺的内容必须在正文中得到兑现,否则会导致用户流失。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-2", "06-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-MC-0155", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "从长篇素材中提取爆款元素时,应从数据冲击、权威背书、金句、认知冲突、痛点共鸣五个维度进行拆解与重组。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0248", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "在利用AI进行任务时,应通过“角色设定”、“任务要求”与“约束条件”三个维度来构建高质量的提示词,以确保输出结果的专业性与效率。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-MC-0251", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "在进行深度文风分析时,应区分有意为之的修辞与无意流露的风格,识别表层修辞与深层结构的错位,并重点关注文本内部的矛盾点以揭示深层逻辑。最终目标应是通过六维雷达图量化分析,并逆向生成可供AI直接使用的提示词。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0253", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "有效的文风分析应包含两个维度:一是给出具体的六维雷达图评分与解读,二是生成可供AI直接使用的逆向提示词。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-4", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0305", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "在处理大量既有素材时,应采取“非破坏性”的提取策略:即在原始素材库之外新建文件夹或输出文件结构,将提取出的观点与案例存储在独立路径中,以防干扰原始文档。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-atomize", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0306", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "在利用现有素材进行内容创作时,应建立独立的素材库结构,通过新建文档进行观点与案例的提取,以确保原始素材文件的完整性与原始文档不被干扰。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "06-9"], "skills": ["xy-ai-workflow", "xy-atomize", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MC-0331", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "在优化AI生成的技能或指令时,应通过明确用户风格、区分口语与书面语、优先展示结果以及使用动态案例等维度进行评估。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MD-021", "type": "method", "confidence": "high", "topics": ["内容创作与平台"], "knowledge": "一鱼多吃不是复制粘贴而是重新表达:同一个核心观点,深度平台给完整论证、种草平台给清晰利益点、短视频给情绪钩子、社交平台给犀利观点加数据。复制粘贴会被判低质,重新表达本身是在训练理解深度。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-10", "06-9"], "skills": ["xy-ai-workflow", "xy-clip", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-MD-023", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "「方法的方法」两步:①相信一个博主的方法论一定藏在他的爆款里(作品比语言诚实,没被结果验证的方法不值得学);②把爆款的文稿、标题、封面分类整理发给 AI,给定明确的提炼方向——底层心理触发机制、认知偏差、传播学、营销心理学、注意力经济——让它逐条列出机制、配原文例句、说明怎么迁移。方向词一变,AI 的输出天差地别。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "06-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-echo-test", "xy-opener", "xy-peer-pick"], "related": null, "q": 2}
{"id": "XY-MD-029", "type": "insight", "confidence": "high", "topics": ["AI与工具"], "knowledge": "大模型的本质是一个搜索机器人:它擅长从海量素材里检索并重组,不擅长凭空创造。所以用 AI 的第一步永远是建自己的素材库;知识库不是存资料的地方而是回答问题的系统,每传一批内容就同步维护索引(有哪些观点、对应什么问题、引用时怎么定位)。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "12-3"], "skills": ["xy-ai-workflow", "xy-atomize", "xy-ip", "xy-vault"], "related": null, "q": 2}
{"id": "XY-MD-031", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "规则密度决定 AI 产出质量:对 AI 的约束越细(字数、句式、分布比例、禁忌、示例)产出越接近你的标准,约束越粗越平庸。「帮我写个标题」然后抱怨 AI 不行,问题不在 AI 在你没给足规则。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["07-2", "12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch", "xy-xhs-headline"], "related": null, "q": 2}
{"id": "XY-MD-032", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "让 AI 产出「像个人」要过三层:语言层不用「您/建议/综上所述」这类教科书腔;结构层别平铺直叙,要有悬念、冲突、对比;灵魂层必须注入你的经历、观点、态度——这是 AI 无法凭空生成的部分。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 2}
{"id": "XY-IPB-0129", "type": "method", "confidence": "high", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "提取成功因子的标准化流程:1. 通过AI/数据获取线索;2. 进行AB Test(灰度验证)以剔除干扰变量,确认单一因子的真实贡献;3. 结合薪酬绩效进行规模化落地。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "11-7"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-IPB-0148", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "高效使用AI(如GPT)的提示词框架:1. 交代背景(赋予AI身份与上下文信息);2. 明确目的(明确任务类型、数量及目标受众);3. 详细描述需求(包含具体数据、形容词、对象,避免啰嗦但要精准)。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-IPB-0255", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "AI在内容生产领域的降本增效逻辑:利用‘流程编排智能体’替代重复性的文案拆解、视频制作等人工任务。目标不是单纯裁员,而是通过‘操盘手的脑子 + AI的手’,实现单人完成原先整个团队的工作量。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "06-10"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-IPB-0259", "type": "anti-pattern", "confidence": "high", "topics": ["私域运营", "流量获取", "AI与工具"], "knowledge": "在客服/私域应答场景中,单纯依赖通用大模型可能导致用户体验下降。原因在于:1. 无法处理复杂的情绪波动(如用户愤怒时的非理性表达);2. 无法处理长上下文的碎片化问题(用户可能需要多次对话才能表达完整诉求)。若AI无法理解上下文或响应过慢,会加剧用户愤怒而非降本提效。因此,必须强化上下文训练和场景化的逻辑处理。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["08-4", "05-3"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-IPB-0299", "type": "method", "confidence": "high", "topics": ["IP人设", "AI与工具"], "knowledge": "某博主分享的“超级个体AI工具组合”方法论:\n1. Notion:作为知识库与大脑延伸,负责信息的存储、积累与整理。\n2. GPT:作为7×24小时的个人助理,负责输出清晰思路、整理混乱信息并提供情绪价值。\n3. DeepSeek:作为思考伙伴,在特定语境下进行逻辑碰撞与深度思考。\n4. Perplexity:作为研究伙伴,负责获取精准、专业的行业洞察与前沿资料。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "13-1"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-IPB-0330", "type": "method", "confidence": "high", "topics": ["认知与心态", "AI与工具"], "knowledge": "在AI时代保持清醒的三个生存法则:\n1. 建立“数据溯源”本能:识别信息来源的质量,警惕低质量社交媒体语料。\n2. 掌握“反刍验证”技能:利用不同AI模型进行交叉验证(如将GPT的答案喂给Claude进行纠错),人工设置检验点。\n3. 打造“物理认知锚点”:通过阅读纸质书、与一线专家面对面交流、保持手写习惯等方式,对抗数字世界的认知退化。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-7", "12-2"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-IPB-0331", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "掌握“反刍验证”技能,通过不同AI模型进行交叉验证。例如将一个AI生成的答案喂给另一个AI并询问其错误之处,或人工设置检验点,至少交叉验证三个信源,以防陷入单一模型的逻辑陷阱。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-IPB-0393", "type": "method", "confidence": "high", "topics": ["IP人设", "AI与工具"], "knowledge": "实现AI持续进化的‘反馈循环’方法:1. 喂入大量结构化数据进行初步训练;2. 在固定问题上让AI给出答复;3. 人类通过人工干预(指出哪些观点不对、哪些地方需要修正)进行纠偏。这种‘像教孩子一样’的持续修正过程,能让AI模型不断逼近人类专家的水平。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-IPB-0413", "type": "method", "confidence": "high", "topics": ["IP人设", "AI与工具"], "knowledge": "在利用 AI 模拟人工角色时,提示词工程应包含两个核心维度:1. 角色定义(Role),确定岗位职责;2. 语言风格(Style),确定表达方式。这决定了 AI 是否能像真人一样沟通。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-IPB-0513", "type": "method", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "AI创业成功的关键路径在于:1. 快速验证场景与流量能力;2. 通过场景套用大模型完成收入层面的MVP(最小可行性产品)测试;3. 利用产生的收入和用户规模数据来优化模型、说服投资人并进行规模化扩张。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "11-8"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-BKG-327", "type": "case", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "讲自己做的项目时主动说出具体的失败细节和金额(「响应确实快,但口袋真的受不了」「一晚上——145美金」),而不是只展示成功结果,这种带具体数字的「翻车自曝」是判断内容是否真人感的强信号——AI生成的产品介绍类文案几乎不会主动暴露成本失控这种负面细节。该条播放3.3万,收藏187,属于同批长文字说明类内容里表现较好的一条。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-2", "13-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 2}
{"id": "XY-BKG-329", "type": "case", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "遇到自己也搞不懂的技术细节时,真实反应是「打电话问另一个AI」这种带点荒诞感的具体行为,而不是自己装懂讲一堆专业术语;这种「承认无知+具体求助动作」的细节比空泛的「我研究了一下」更有真人感,也更容易让观众会心一笑。该条播放2.68万。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-2", "12-1"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 2}
{"id": "XY-BKG-601", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "开头用连环追问轰炸一串专业名词制造「我听不懂但很想懂」的焦虑,再立刻承诺一次性讲清楚——密集的疑问句本身就是钩子,因为观众会下意识想验证自己到底认不认识这些词;某企业AI科普视频用这个结构开头,播放38万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-BKG-602", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "开头先转述一个别人的反常识论断,自己当场表演「被冒犯到反驳」的反应(比如脱口而出「扯淡吧」),再回头解释对方其实说得对——这种自我反转的表演比直接陈述观点更能勾住观众的好奇心,因为观众想知道到底是谁对;某AI行业视频用这个结构,播放136万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "13-1"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-BKG-605", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "开头用一个具体的、发生在自己身上的场景做钩子(刷到什么、看了多久没划走),而不是直接讲主题本身——先让观众相信讲述者自己也是被内容吸引住的普通观众,再顺势带出话题,比直接说「今天我们来聊xx」更有代入感;某AI工具科普视频用这个结构开头,播放275万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-2", "13-1"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-BKG-611", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "开头先否定一个大众默认的行业共识(现在市面上90%的xx其实都是伪命题),再给出自己的重新定义——先破后立的结构比直接讲定义更容易让观众产生原来我一直理解错了的认知落差感;某企业AI科普视频用这个开头,播放38万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "13-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-BKG-613", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "全篇用同一个具象隐喻(把AI比作养的一只虾/新员工)反复呼应,每讲一个新知识点都套回这个隐喻框架里(养灵魂、养记忆、教技能),观众不需要记住抽象的技术概念,只需要记住一个故事线就能跟住全篇节奏,比纯知识点罗列更抗划走;某AI协作方法论视频用了这个结构,播放275万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "12-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-script-glue"], "related": null, "q": 2}
{"id": "XY-BKG-615", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "长篇实操演示类视频里如果中途真的遇到报错/翻车,不剪掉、顺势讲成debug教学环节——真实的意外比设计好的演示更能维持观众注意力,因为观众会好奇这下要怎么解决,天然形成一次小高潮,比全程顺风顺水的演示更抗划走;某Claude Code入门教程视频用这个结构,播放304万、收藏128万(收藏率超4%)", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["13-2", "06-8"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-BKG-619", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "讲解抽象技术概念时,全篇统一用一个类比框架(把AI大模型比作新员工,把提示词比作工作指令,把知识库比作入职培训资料),每个新名词出现都套进这同一个比喻体系里解释,观众不需要切换理解框架,跟随节奏明显更顺;某企业AI科普视频用这个结构,播放38万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-2", "07-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-BKG-623", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "把观众明确划入一个你已经比大多数人领先的身份群体(你现在能刷到这条视频,证明你已经是站在最早期的这批人了),用身份认同和优越感做共鸣钩子,比单纯讲行业机会更能激发点赞和转发——因为转发的动机从分享信息变成了证明自己是这批人;某AI行业趋势视频用这个话术,播放137万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-3", "01-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-content-scan", "xy-opener"], "related": null, "q": 2}
{"id": "XY-BKG-624", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "把一个遥远的宏观数字(几千亿美元的行业投资)翻译成这笔钱最终从哪里来的具体个人命运叙事(裁员、工资单),用这跟我有什么关系的具体化处理把行业新闻变成个人焦虑,是这类内容能引发大量转发提醒他人的核心机制;某讨论AI资本逻辑的视频用这个结构,播放17万、评论率2.7%(远高于同批平均)", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["01-4", "11-12"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-BKG-631", "type": "insight", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "用少即是多的反潮流定位(别人都在拼命加功能,我反而把功能做到最少)制造认知优越感式共鸣——观众会因为发现了一个和大众选择不一样但更聪明的方案而产生分享欲,这种我发现了一个内行秘密的心理比单纯介绍产品功能更能驱动转发;某AI工具测评视频用这个结构介绍一款小众工具,播放83万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip"], "related": null, "q": 2}
{"id": "XY-BKG-632", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "用可验证的具体操作时长(144秒、58秒、101秒完成一个应用)反复作为证据锚点,而不是笼统说很快——精确到秒的数字比模糊的形容词更有说服力,也更容易被观众截图转发当谈资,是技术类内容建立可信度的常见手法;某AI建站工具测评视频反复用这个手法,播放740万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-BKG-633", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "口播文案里保留大量「啊」「呢」「呗」「对不对」这类语气助词和高频反问,虽然从书面语角度看是冗余的,但恰恰是这些冗余制造了说话人在跟你实时对话的临场感,是真人感的重要来源——过度精简、每句话都信息密度拉满、没有一个语气词的文案反而更接近AI生成的书面语感;多条高播放量口播类视频都保留了这种口语填充", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-3", "06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 2}
{"id": "XY-BKG-634", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "情绪激动时出现的重复和不流畅表达(连续的你你你你、兄弟反复称呼)没有被剪辑打磨掉,反而增强了真实感——AI生成的文案通常语法工整、逻辑顺滑、不会有这种话赶话说不利索的痕迹,保留这种口误感是判断一段文案是不是人味的直观标志之一", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-3", "06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 2}
{"id": "XY-BKG-635", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "在视频结尾打破镜头前完美讲述者的人设,插入一段和镜头外编导/团队成员的真实拌嘴(对方问了个问题,讲述者当场怼回去),这种没有被剪掉的幕后互动比任何精心设计的花絮更能证明这是一个真实的人在真实的工作场景里,是human-touch的高强度信号;某商业方法论视频用这个结尾,播放51万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-2", "13-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch", "xy-ip"], "related": null, "q": 2}
{"id": "XY-BKG-636", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "情绪自然溢出时直接使用口语化的粗口式表达(而不是刻意设计的博眼球脏话),是判断一段文案是不是现场真讲而非逐字念稿/AI生成的信号之一——AI生成的文案几乎不会主动带脏字,因为脏字通常伴随着讲述者当下真实的情绪波动,无法提前设计;出现在某IP方法论分享视频里,播放9万", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-7", "07-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip"], "related": null, "q": 2}
{"id": "XY-BKG-639", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "口播中出现带不确定性的推测语气(我猜测它呢可能内置了……),而不是对不了解的技术原理强行给出确定性的解释,这种知之为知之不知为不知的坦白反而增加了可信度——AI生成的解说词更容易为了显得权威而把猜测包装成确定性陈述,保留「我猜」「大概」这类限定词是真人感的一个细节标志", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-3", "07-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 2}
{"id": "XY-BKG-905", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "持续收集「亲和性赛道」(跟自己主赛道人群重叠但不是同一个赛道,比如自媒体知识付费人群同时也爱看商业思维、AI科技类内容)里表现好的爆款开头,把开头的句式结构抽出来,替换成自己业务的内容去用——因为跟自己不是同一个赛道,就不构成抄袭。作者自己用这套方法写出的一条开头做到过5秒完播率55%。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-1", "03-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-opener", "xy-peer-pick"], "related": null, "q": 2}
{"id": "XY-BKG-931", "type": "case", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "一段现场采访的原始转写,充满语气词、被打断的短句、答非所问再纠正的真实对话痕迹,没有被剪成一段流畅的独白稿。这类未经过度打磨、保留了真实交流节奏(停顿、追问、纠正)的内容,天然比逐字稿念白更像一个真实的人在说话——判断「活人感」的一个可复制标准,不是语言有没有瑕疵,而是有没有被剪辑抹平所有真实反应的痕迹。这段采访播放102175。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-3", "06-7"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-BKG-935", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "三类内容对访问主页的转化效果差异很大:讲反常识判断/观点的比纯讲方法的主页访问率高90%;讲自己思考过程和亲身经历的自我披露型内容,比纯讲理论的主页访问率高60%。观众想认识的是一个「人」而不是一套「知识」,暴露越多真实的思考轨迹和个人经历,内容背后的这个人越显得可信、越让人想点进主页多看看。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-2", "08-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-human-touch"], "related": null, "q": 2}
{"id": "XY-BKG-936", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "「晒素材」被作者认为是最强的一种创作手法,但也最难模仿——因为它不是编一套话术证明自己厉害,而是把真实做过的事情原样讲出来(比如把方法论和工具箱免费公开,然后靠这件事间接变现了七位数)。核心难点不在文案技巧,而在于必须真的先把这件事做完;观众信的不是你会不会说,是你有没有真的做过。这套内容前后两次发布,播放分别为16673和26291,数据稳定。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-4", "07-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-content-scan", "xy-human-touch", "xy-peer-pick"], "related": null, "q": 2}
{"id": "XY-MB-009", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "过度依赖AI,对本来就不擅长内容的人来说是一种「资源诅咒」——如果没有AI,反而会更早认清自己不擅长、更早去学习「笨办法」打磨基本功;有了AI之后,容易一直停留在研究提示词技巧的层面,迟迟不去补最根本的短板。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "06-11"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-MB-321", "type": "anti-pattern", "confidence": "medium", "topics": ["团队与模式设计", "认知与心态", "AI与工具"], "knowledge": "“AI时代人人可以当一人公司”的潜台词其实是一种幻觉:认为只要有了免费的AI劳动力,自己不需要改变、不需要提升能力就能开始赚钱。但实际情况是:不会运营的人,即便给他无限量的免费AI产出,依然无法把账号做起来;真正会运营的人,AI只是放大器,而不是能力的替代品", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "06-10"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-MB-339", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "真人出镜没流量时想靠数字人出镜、真人配音没流量时想靠AI配音、自己一条视频都没火过却想靠AI批量制造爆款——这个逻辑本末倒置:人工把单条内容做好,是机器批量复制的前提,而不是反过来。机器做内容比人工难10倍,机器批量做内容比人工难100倍;如果连以利润为前提去衡量产出,难度还要再乘一个数量级——自己做不好的事情,交给机器做也做不好", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-11", "06-10"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-content-scan", "xy-human-touch", "xy-slow-lane"], "related": null, "q": 2}
{"id": "XY-MB-357", "type": "method", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "想让 AI 模仿自己的写作风格,前提是能明确地把自己的风格拆解成具体参数:常用词汇选择、句式结构、语法特点、修辞手法、语气语调、情感表达、叙述视角、主题偏好、意象比喻、节奏韵律、逻辑结构、语言繁简程度等。没有先做这层拆解,就抱怨大模型模仿不像,是不成立的。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-peer-pick", "xy-term-crack"], "related": null, "q": 2}
{"id": "XY-MB-358", "type": "anti-pattern", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "用大模型模仿'有认知'的表达时,容易默认套用'反问+感叹'的句式(例如'你以为这样就能成功?错!'),营造出说教感,误以为这样显得有认知。这种表达方式应当在提示词中被明确禁用,包括禁用感叹号。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-7"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-peer-pick", "xy-term-crack"], "related": null, "q": 2}
{"id": "XY-MB-389", "type": "anti-pattern", "confidence": "medium", "topics": ["选品逻辑", "合规与风控", "AI与工具"], "knowledge": "真正有用、好用的工具/智能体往往卖不掉,看起来有用但实际没什么用的'魔法感'工具反而更容易卖掉。前者虽然有用,但使用频率不高,很难成为高频刚需产品;后者的核心是让不懂行的人产生'有了这个黑魔法我就能做成事'的错觉,一旦这个营销卖点成立,产品本身是否真正有用反而不再重要。这种反差值得从业者警惕,避免不自觉地滑向后者的套路。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "01-5"], "skills": ["xy-ai-workflow", "xy-publish-guard", "xy-selection"], "related": null, "q": 2}
{"id": "XY-MB-393", "type": "method", "confidence": "medium", "topics": ["选品逻辑", "新人起步方法论", "AI与工具"], "knowledge": "面向中国大众/下沉市场设计 AI 或知识类产品时,几个关键的接地气要求是:不需要额外的网络门槛就能访问;最好有轻量入口或至少能在常用聊天软件内打开;支持手机号或第三方账号登录,而不是需要国外账号体系登录;能让用户免费体验到较强的模型能力,而不是每一步都要额外付费;操作路径要足够简单,减少不必要的按钮和步骤。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["03-8", "12-2"], "skills": ["xy-ai-workflow", "xy-newbie", "xy-selection"], "related": null, "q": 2}
{"id": "XY-MB-436", "type": "method", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "让AI产出好文案有一条红线:提示词里绝对不能出现固定的模板、公式(比如固定的'爆款叙事模板');正确的做法是持续告诉AI什么是好、什么是坏,定义清楚判断规则,然后让AI在规则内自由发挥——如果AI依然产不出好内容,问题出在规则不够完善,而不是这条路线本身错了", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MB-439", "type": "anti-pattern", "confidence": "medium", "topics": ["选品逻辑", "认知与心态", "AI与工具"], "knowledge": "'这个产品没必要买,AI都能直接生成'这句话,只适用于自己作为消费者时的判断,不适用于创业做生意时的判断;觉得'没必要买',是因为自己已经知道用哪个模型、哪个产品、配合什么提示词能做出来——但信息差永远存在,产品被AI复刻的门槛有多低,都不能证明这个产品不能赚钱", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-selection"], "related": null, "q": 2}
{"id": "XY-MB-533", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "我见过太多人指望AI代替自己把内容做起来,但翻遍案例:有没有一个人是自己本来做不出好内容,靠AI代替自己就真的拿到结果的?不分国内国外、图文视频、男女老少,我是真没见过——AI放大的是你本来就有的能力,不是从0生出能力。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "06-10"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MB-544", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "内容创作与平台", "AI与工具"], "knowledge": "很多人用AI写不出好文案,第一反应是我不懂AI或者我没选对AI产品——这其实是一种自我保护式的归因,真正的原因往往是自己对什么是好文案都没有判断力,换十个AI产品也一样写不好。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-3", "12-3"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-MA-1575", "type": "case", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "据案例描述,一位内容创作者兼商业顾问有个人IP内容、客户咨询交付、线上课程、团队管理 4 个业务板块,另一位知识IP有短视频、私域运营、图书出版、线下课程、企业咨询、方法论沉淀 6 个板块,两者都用同一套“想法工作台+通用知识库+业务板块”结构,只是业务板块按各自职责定制。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["13-1"], "skills": ["xy-ai-workflow", "xy-idea-desk", "xy-ip", "xy-mode", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-MC-0083", "type": "method", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "知识的内化过程存在两种路径:一是从默会知识向显性知识转化(即“我知道但说不清楚”到“我能讲清楚”);二是向隐性知识转化(即“别人讲清楚了但我不会”到“我学会了”)。", "source_type": "adapted_course", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "06-10"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-BK3-0100", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "“貌似合理”陷阱:在商业决策或营销文案中,过度追求描述的“连贯性”和“故事感”,容易误导判断。当一个方案被描述得极其详尽、细节丰富(例如:某产品不仅能解决问题,还能通过AI算法实现精准预测并提供24小时人工响应)时,这种“丰富性”会给人一种极高的可信度错觉。但生意操盘手必须警惕:细节的堆砌往往会增加执行层面的不确定性,越是看似“完美且有条理”的描述,其落地成功的真实概率往往比简单直接的方案更低。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-6"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-content-scan"], "related": null, "q": 2}
{"id": "XY-BK3-0141", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "技能错觉(Skill Illusion):指专业人士因掌握了大量知识,容易产生“我能掌控未来”的虚假自信。在商业实战中,这表现为过度依赖过往成功经验来预测未来趋势(后见之明),忽视了环境的动态变化。判断标准:当一个人在面对不确定性时,表现出的自信程度与他所掌握的知识深度成正比,而非与环境的复杂程度成反比时,即存在技能错觉。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-6"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-BK3-2053", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "知识过时风险与“停滞期”陷阱:传统技术(如手工艺)具有长效性,而知识型技能的半衰期极短。如果业务模式过度依赖于某种特定知识或单一技术路径,一旦该领域发生范式转移(如AI取代基础分析),企业将面临毁灭性打击。此外,知识工作者在达到一定成就后容易进入“停滞状态”,若不建立非竞争性的兴趣或持续学习机制,其创造力会迅速枯竭。在团队管理中,需警惕员工因过度专业化而导致的思维僵化。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-10", "06-10"], "skills": ["xy-ai-workflow", "xy-coach"], "related": ["XY-CS-0858"], "q": 2}
{"id": "XY-MSK-108", "type": "principle", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "“技术融合度”决定了控制权。当一项颠覆性技术(如AI)的进化速度远超人类的理解与响应带宽时,该技术会逐渐脱离控制。若无法实现人机(或管理者与执行层)的高度融合,技术/组织就会演变为“异类”,产生反噬。生意场景中:在数字化转型或引入自动化工具时,如果业务逻辑与技术实现之间缺乏深度融合(沟通带宽低),系统将无法执行人的意志,反而会产生不可控的副作用。", "source_type": "adapted_book", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "11-6"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": ["XY-CE-1019"], "q": 2}
{"id": "XY-KSE-0008", "type": "method", "confidence": "medium", "topics": ["IP人设", "AI与工具"], "knowledge": "议程驱动型选题法:通过先定“核心判断(议程)”再进行内容生产的顺序,实现选题的可批量化生成。一旦确定了要传递的核心价值观或专业判断,就可以利用AI或思维导图将该议程拆解为无数个子议题。在生意场景中,这要求老板或操盘手在拍摄前必须完成“定调”动作,而不是拍完一堆零散内容后再去分类。这种方法能确保所有输出的内容都在不断重复同一个“点”,通过高频次的认知占领,降低用户的决策成本。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": null, "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-roundtable", "xy-selection"], "related": ["XY-BK3-4318", "XY-CE-0522"], "q": 2}
{"id": "XY-CE-0209", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "客户赋能(Client Enablement)方法:不仅仅是交付一份报告,而是通过与客户团队深度协作、技能转移(Skill Transfer),确保战略方案能够从纸面落到业务执行层面。在生意场景中,这意味着服务商不应只做“外脑”,而要通过带教、流程再置和组织架构设计,让客户的内部团队具备独立执行该战略的能力。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["09-5", "12-3"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-1069", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "技术革命驱动的产业生态位迁移:人类社会正经历以数字化、智能化、绿色化为核心的第四次技术革命。在大数据成为最核心资产(预计到2030年贡献约45%经济价值)的背景下,企业必须意识到技术变革会重塑生产力结构(如AI在2025年贡献1.7万亿美元产值)。对于生意操盘手而言,判断一个赛道的潜力不仅要看当前利润,更要研判其在技术演进周期中的生态位。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-4", "13-3"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CE-1246", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "数字化生态构建法(Digital Ecosystem Building)。领先的跨国药企已不再仅关注单一产品的销售,而是通过建立“数字化创新中心”或“AI智能平台”,与本土初创公司、科技巨头合作,构建以患者为中心的生态系统。这包括在线咨询、处方续药(Rx refilling)、药品配送及全周期的疾病管理。对于私域生意而言,这提示了从“卖产品”向“提供全流程服务方案”转型的必然性。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-1", "08-7"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-1285", "type": "principle", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "数字化运营模式转型(Digital Operating Model)强调组织必须从传统的线性管理转向数据驱动的决策模式。这包括:1. 引入AI与大数据进行预测性分析;2. 构建能够快速响应市场变化的敏捷团队;3. 建立闭环的评估体系。在生意场景中,这意味着如果你的私域运营还在靠“拍脑袋”决定发什么券、什么时候做活动,就无法应对竞争。必须建立“数据采集-分析建议-执行反馈-实时调整”的闭环,利用数字化工具实现规模化、自动化的客户支持。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "11-6"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-1451", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "单一增长依赖陷阱:如果产业结构中某一环节(如传统化学药)占比过高且增速放缓,整体增长将面临瓶颈。在生意决策中,必须警惕“现金牛”业务的衰退风险,应提前布局具有高增长潜力的“新动力”领域(如AI+医疗、数字化服务),实现从“稳定基础”向“增长引擎”的结构性升级。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-10", "01-4"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CE-1991", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "合规与风控", "AI与工具"], "knowledge": "业务规模化中的“技术债”与“人才依赖”风险。随着企业试图通过AI或自动化手段提升效率(如实现跨国/跨品类的个性化推荐),会面临三大核心挑战:一是前期高额的软硬件投入与对具备算法能力的专业人才的依赖;二是过度依赖第三方供应商可能导致核心技术自主权缺失;三是快速迭代的市场环境下,缺乏明确的维护责任与合规流程。对于生意操盘手而言,在追求自动化带来的效率提升(如销售额10-20%的增长)时,必须同步建立数据治理与合规标准,防止因算法黑盒或隐私风险导致业务崩盘。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-3", "08-9"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-playbook", "xy-publish-guard"], "related": null, "q": 2}
{"id": "XY-CE-2005", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "过度依赖自动化工具陷阱:在业务规模扩张时,企业容易产生“只要引入了AI/低代码工具就能解决所有问题”的错觉。实际上,过度依赖自动化测试或代码审查会导致严重的质量隐患,因为人类员工往往会停止对自动化输出结果的深度校验。在私域运营或业务流程数字化中,如果完全放弃人工对自动化脚本、AI生成文案或低代码逻辑的复核,随着业务量级(Scale)提升,错误会被指数级放大。判断标准:当你的团队开始盲目信任工具输出而减少人工复核环节时,风险正在积聚。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "06-8"], "skills": ["xy-ai-workflow", "xy-ops", "xy-private-ops", "xy-question-spec"], "related": null, "q": 2}
{"id": "XY-CE-2014", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "利基场景验证法(Niche Optimization Method):针对高复杂度、高价值的特定问题进行专项突破。例如通过提高5%的分子模拟精度来开发新产品,或利用更精准的数据采样使AI算法输出提升20%。对于生意人而言,不要试图用单一新工具解决所有问题,而应寻找那些“现有方案效率遇到瓶颈”且“容错率较高”的特定环节(Niche),通过局部效率的微增带动整体利润空间的提升。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-6", "12-1"], "skills": ["xy-ai-workflow", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-2025", "type": "anti-pattern", "confidence": "medium", "topics": ["合规与风控", "AI与工具"], "knowledge": "数据隐私与公平性的冲突陷阱(Privacy-Fairness Tension)。在利用AI模型进行用户画像分析和精准营销时,操盘手常面临两难:一方面为了提升转化率需要收集更多人口统计学数据,另一方面过度采集又会触碰隐私红线并引发合规风险。如果为了追求极致的“用户理解”而过度挖掘敏感数据,不仅会面临监管风险,还会因为缺乏透明度导致用户信任崩塌。在私域生意中,应避免“数据贪婪型”运营,转而采用隐私工程(Privacy Engineering)思维,在保护用户敏感信息的前提下,通过脱敏后的行为数据来驱动业务增长。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-8", "08-9"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-publish-guard"], "related": ["XY-CE-2207"], "q": 2}
{"id": "XY-CE-2186", "type": "case", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "自动化测试与部署的效率革命:通过引入AI驱动的自动化工具,可以实现开发效能的量级提升。在实际业务场景中,例如某大型零售企业通过“基础设施即代码”(IaC)技术,将原本需要数周的部署时间缩短至30分钟;或某金融机构利用AI工具进行单元测试,使核心后端应用的测试编写速度提升了180倍。对于中小规模的生意操盘者,这提示了一个核心逻辑:不要试图通过堆人力来解决重复性的流程问题,而应寻找能实现“自动化部署”和“自动监控维护”的标准化路径,从而将精力释放到更复杂的业务逻辑设计中。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "11-6"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-CE-2352", "type": "method", "confidence": "medium", "topics": ["私域运营", "流量获取", "AI与工具"], "knowledge": "样本池化与效能提升法(Sample Pooling for Efficiency):在资源有限或需要快速大规模覆盖时,应采用“样本池化”逻辑来提升处理效率。即通过对多个个体进行初步的合并检测/筛选,快速排除大量“无风险”样本,从而将有限的专业资源(如高成本的人工服务、深度运营精力)集中在“疑似/高价值”样本上。在生意场景中,这对应于漏斗式筛选逻辑:先用低成本的自动化工具(如AI客服、标准化问卷)进行大规模初筛,再将筛选出的高意向客户转交给高成本的人工专家(如金牌销售)进行深度转化,以实现规模化与精准度的平衡。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["06-11", "09-1"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-2733", "type": "method", "confidence": "medium", "topics": ["私域运营", "流量获取", "AI与工具"], "knowledge": "AI驱动的规模化个性化(Scalable Hyper-personalization):利用生成式AI技术解决“触达范围与服务深度”的矛盾。传统模式下,极致的1:1定制化只能服务于极少数VIC(超级重要客户),而通过AI工具,可以将这种高标准的客户体验(CX)和个性化服务逻辑,低成本地延伸至中层及渴望型客户(占奢侈品客群90%以上)。在私域场景下,这意味着利用AI提升导购的专业度与响应速度,实现大规模、标准化的精细化运营。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "13-1"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-2810", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "数字化转型的三大核心价值维度:企业通过引入颠覆性技术,应重点关注三个维度的转型:1. 运营效率提升(Operational Effectiveness),通过自动化和AI优化现有流程;2. 商业模式适配(Business Model Adaptation),捕捉价值转移的新增长点;3. 全链路数字化(End-to-end Digitization),实现从研发、采购到销售的全流程数据打通。对于生意操盘手而言,数字化不是为了买软件,而是为了通过数据驱动决策,实现从单点优化到全链路协同的跨越。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-1", "12-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-goal-card", "xy-mode"], "related": null, "q": 2}
{"id": "XY-CE-2811", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "高价值数字化应用优先级模型:在资源有限的情况下,应优先投入具有高回报潜力的五个领域:1. 端到端动态供应链优化(从供应商到终端客户);2. 基于数据的预测性维护(降低设备/库存损耗);3. 手动流程的自动化与数字化工作流;4. 实时数字绩效管理;5. 通过AI实现产量/能效优化。在私域运营中,这对应着优先做“全链路数据打通”和“自动化成交路径”,而非碎片化的工具堆砌。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8", "12-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-private-ops", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-3155", "type": "method", "confidence": "medium", "topics": ["私域运营", "流量获取", "AI与工具"], "knowledge": "基于AIPL模型(认知、兴趣、购买、忠诚)的流量转化路径优化:生意增长由四个维度驱动。在获客阶段,利用大促节点(如阿里超级日)或媒体投放获取流量;在转化阶段,通过折扣和促销手段提升购买率;在留存阶段,利用短信、小程序等私域工具进行直接客户互动(Direct Customer Engagement),通过会员计划提升复购频率。生意操盘手应根据当前品牌所处的AIPL阶段,匹配对应的投入杠杆(ATL媒体、BTL促销或私域互动)。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["03-7", "08-6"], "skills": ["xy-ai-workflow", "xy-close", "xy-ip", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-3472", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "模块化架构解耦:通过引入中间件(如EAI)或标准化接口,将前端用户体验与后端复杂的业务逻辑进行解耦。在生意场景中,这意味着不要让前端App或小程序承载过重的业务计算逻辑,而是通过后端标准化模块处理交易,实现“前端轻量化、后端集中化”,从而提升多渠道(Web/App/小程序)的体验一致性与维护效率。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-11"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-4518", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "人才盘点与动态退出机制。通过绩效管理识别高绩效员工(High Performers)与技能过时员工(Obsolete Skills Employees),并实施差异化策略:对高绩效者进行晋升或激励,对技能过时者进行重新培训(Retrain)或有序退出(Exit)。在生意场景中,这要求老板具备清晰的岗位能力模型,不能让低效员工长期占据核心资源位。通过这种方式可以优化人力成本结构(FTE Savings),实现从“规模扩张”向“人效提升”的转型。判断标准:当团队整体产出遇到瓶颈,且发现大量人力投入在低价值或重复性工作中时,必须启动人才盘点。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": ["XY-CE-4538"], "q": 2}
{"id": "XY-CE-4538", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "人才盘点与技能迭代法:通过绩效管理识别高绩效员工(High Performers)与技能过时员工(Obsolete Skills)。在生意规模扩张或业务转型期,应建立“晋升+重塑”机制:对核心人才进行激励与提拔,对技能落后的员工实施再培训(Retrain)或果断退出(Exit)。这能避免团队陷入“平庸陷阱”,确保组织人力资源与业务战略目标高度匹配。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "11-11"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": ["XY-CE-4518"], "q": 2}
{"id": "XY-CE-4591", "type": "method", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "变革驱动力模型(意愿/技能/紧迫性):评估一个业务转型或新项目落地难度时,需从三个维度进行诊断。1. 紧迫性(Urgency):业务是否面临生存危机?2. 意愿(Willingness):高层与一线是否达成共识并愿意改变?3. 技能(Skill):组织是否具备实现目标所需的专业能力。在生意实战中,若“紧迫性”极高但“技能”缺失,必须引入外部专家进行“并肩作战式”辅导;若“意愿”极高但“紧迫性”不足,则需警惕员工因过度兴奋而导致的后期疲软。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8"], "skills": ["xy-ai-workflow", "xy-question-spec"], "related": null, "q": 2}
{"id": "XY-CE-5248", "type": "method", "confidence": "medium", "topics": ["选品逻辑", "AI与工具"], "knowledge": "“高价值赛道筛选模型”:在评估新兴市场或新业务机会时,应从三个维度进行交叉验证。第一是“技术赋能度”,即业务是否利用了先进技术(如AI、远程监控)来替代传统低效模式;第二是“社会价值契合度”,即业务是否切中了人口结构变化带来的刚需(如老龄化下的HealthTech、教育升级);第三是“可持续性/合规度”,即业务是否符合绿色经济与环保标准。对于私域或品牌操盘手,这意味着选品和业务逻辑不能只看短期流量红利,而要看该行业是否正处于从“高速度”向“高质量”转型的结构性红利期。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-3", "12-1"], "skills": ["xy-ai-workflow", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-5474", "type": "principle", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "数字化转型KPI双维度设定原则:实施自动化或AI工具时,目标必须同时涵盖“业务增长”与“运营效率”两个维度。在生意场景中,不能仅盯着成本降低(如FTE减少30%),还必须挂钩业务增量指标(如客户渗透率提升30%、交叉销售倍数增加2.5x),防止为了单纯的降本而损害了客户体验或业务规模。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-7"], "skills": ["xy-ai-workflow", "xy-goal-card"], "related": null, "q": 2}
{"id": "XY-CE-5532", "type": "method", "confidence": "medium", "topics": ["私域运营", "团队与模式设计", "AI与工具"], "knowledge": "“生成式AI驱动的运营全链路提效”:Gen AI(如ChatGPT)已从单纯的对话工具进化为消费品运营的基础设施。在实际生意场景中,AI的应用不应仅停留在写文案层面,而应深度嵌入到:产品描述自动化、个性化营销内容生成、选品规划(Assortment Planning)、供应链优化以及消费者情绪分析等核心环节。对于中小规模的生意人,利用AI进行“设计灵感”和“消费者情绪监测”是实现低成本、高频次市场反馈的关键手段。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "06-11"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-5877", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "预测性维护与行为预判(Predictive Maintenance & Behavior Prediction):利用AI/ML技术对设备状态或用户行为进行建模,实现从“事后维修”到“事前预防”的转变。在生意场景中,这意味着通过分析历史数据(如设备故障频率、客户复购周期)来提前介入。例如,在制造业中预测设备何时需要保养以避免停工;在零售私域中,通过分析用户购买频率预测其流失风险或下一次复购时机,从而实现精准的营销介入。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "08-5"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-CE-5880", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "自动化决策与智能助手(AI Digital Assistant & Automated Decisioning):利用深度学习进行实体提取、语音翻译或自动化决策,实现业务流程的智能化。在生意场景中,这体现为从“人工客服”向“智能导购/数字助理”的转型。通过对用户意图的精准识别,实现营销内容的自动化分发和个性化推荐,从而在规模化运营的同时保持极高的用户体验精度。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "09-1"], "skills": ["xy-ai-workflow", "xy-close", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-5892", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "新兴技术预算占比增长逻辑:企业对新兴技术的投入比例正处于爆发前夜,预计在未来4-6年内,新兴技术在IT预算中的占比将从目前的10%增长至30%,实现3倍的规模扩张。这种转变的核心驱动力在于三点:1. 效率提升(通过自动化实现);2. 用户体验增强;3. 产品研发周期的加速。对于生意操盘手而言,这意味着在布局新赛道或引入新技术工具(如AI、自动化营销工具)时,应预判未来3-5年内行业整体预算向此类技术倾斜的趋势,提前完成供应链或技术栈的卡位。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CE-5937", "type": "method", "confidence": "medium", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "技术投资矩阵分类法:通过“通用性(Universal)”、“跨领域增长型(Cross-sectional)”、“垂直行业特定型(Vertical specific)”和“利基市场型(Niche)”四个维度对技术应用进行分层。在生意决策中,应根据自身资源配置重心选择赛道:通用型技术适合做底层标准或平台;跨领域增长型适合寻找多个行业共性的痛点(如AI在零售与医疗的通用性);垂直型技术适合深耕特定行业壁垒(如交通领域的传感器技术)。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-1"], "skills": ["xy-ai-workflow", "xy-mode", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-5944", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "行为分析(Behavioral Analytics)指利用AI/ML等技术从电商或在线平台的消费者行为数据中提取洞察。在私域生意场景下,这不只是看成交额,而是通过分析用户的点击路径、停留时长、互动频率等行为特征,构建用户画像。操盘手应以此判断用户的生命周期阶段(新客、活跃、流失预警),从而实现精准的促活或召回,将“盲目发券”转变为基于行为触发的自动化运营。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["08-3", "12-2"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-5947", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "虚拟代理(Virtual Agents)是通过预设规则和AI技术提供自动化服务或引导的软件程序。在私域运营中,这通常表现为智能客服或自动化导购机器人。操盘手应将其视为“24小时在线的标准化销售”,通过设定标准化的SOP(标准作业程序)和智能问答逻辑,解决80%的重复性咨询问题,从而释放人工精力去处理高客单价或高情感价值的深度沟通。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "05-3"], "skills": ["xy-ai-workflow", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-5950", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "技术滞后与品牌错位风险:在快速迭代的细分领域(如SAIS注射系统),如果技术标准和质量水平落后于头部玩家,即便拥有品牌优势也难以维持市场地位。生意人需警惕“技术代差”带来的生存危机,尤其是在面对具备长期合作经验的本土厂商(如案例中已合作10年的厂商)时,单纯靠品牌溢价无法抵御技术迭代带来的份额流失。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-3", "10-6"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CE-5951", "type": "method", "confidence": "medium", "topics": ["选品逻辑", "AI与工具"], "knowledge": "多利益相关方需求合成法(Stakeholder Needs Synthesis):在复杂商业决策中,不能仅听取单一终端用户的声音。有效的市场评估必须同时覆盖:患者(依从性/便捷性)、医生(临床效果/操作简便性)、护士(使用安全性)以及药企客户(成本控制/合规性)。只有将这些冲突或重叠的需求进行合成,才能识别出真正具备商业价值的产品概念(如案例中提到的重新配置系统和合规解决方案)。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["09-1", "11-1"], "skills": ["xy-ai-workflow", "xy-question-spec", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-5953", "type": "method", "confidence": "medium", "topics": ["选品逻辑", "AI与工具"], "knowledge": "产品形态升级驱动市场渗透:通过改变产品的物理形态(Formulation)来提升用户体验,是打破原有竞争格局的关键手段。案例显示,在类风湿关节炎(RA)领域,从粉剂向液态制剂的转型是推动市场规模增长的核心驱动力之一。生意操盘手在做产品迭代时,不应只盯着核心功能,更要关注“交付形态”是否符合用户使用习惯(如注射笔的便捷性),这是实现从低端市场向高端市场渗透、或从本土品牌向国际标准靠拢的必经之路。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-6", "10-5"], "skills": ["xy-ai-workflow", "xy-ops", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-5956", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "大客户早期介入策略(Early-stage BD Involvement):针对高增长潜力的头部药企,BD部门不应仅作为订单接收方,而应在客户产品研发的早期阶段(如液态制剂开发期)即介入。通过提供共同开发、技术共享及生产协作,将单纯的买卖关系升级为战略合作伙伴关系。在生意场景中,这要求操盘手具备前瞻性的客户画像能力,通过参与对方的研发节奏(Timeline),实现从“卖产品”到“卖解决方案/共同开发”的升维,从而锁定长期订单并建立极高的竞争壁垒。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-4", "09-1"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-mode", "xy-ops", "xy-playbook"], "related": null, "q": 2}
{"id": "XY-CE-5957", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "增值服务驱动型销售(Value-added Service Selling):针对医疗器械或高技术含量产品,单纯的价格竞争是低效的。有效的商业模式应通过提供“全套解决方案”来增强议价能力,包括:本地化设计与制造、市场开发支持(如KOL教育)、以及协助客户进行医保准入的经济学评估。在私域或B2B生意场景中,这意味着你的产品不只是一个SKU,而是包含“产品+技术支持+市场赋能”的综合包,通过解决客户在准入、品牌和生产风险上的痛点来提升客单价与粘性。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-1", "09-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CE-5959", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "单纯的价格/成本导向陷阱(Cost-only Focus Trap):在面对具有高技术壁垒或强品牌需求的目标客户时,如果仅以“低价”作为核心竞争手段,会陷入利润率极低的红海竞争。对于需要高技术迭代(如电子设备、无针注射器)的领域,客户对成本的敏感度相对较低,而对定制化能力、供应灵活性及技术协同的要求极高。如果操盘手过度优化成本而忽视了研发端的深度参与,会错失进入高增长、高利润赛道的门票。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-5", "11-1"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CE-5960", "type": "method", "confidence": "medium", "topics": ["流量获取", "商业案例与实战复盘", "AI与工具"], "knowledge": "市场开发优先级矩阵法:在资源有限的情况下,应根据“营收潜力”与“投资回报率/市场成熟度”两个维度对目标客户进行分层。对于高潜力但低渗透的市场(如糖尿病与类风湿领域),应立即启动市场塑造行动,通过教育患者、建立KOL认知及支持医保准入来“造场”;而对于成熟市场则应侧重于存量维护。生意场景中,这意味着你不能对所有潜在客户平均用力,必须识别出哪些是“高价值且需重点教育”的战略账户,并为其匹配专项预算。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8", "03-7"], "skills": ["xy-ai-workflow", "xy-ops", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-5964", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "碎片化运营陷阱:在进行市场开发或客户管理时,如果仅有零散的市场感知(Good pulse)而缺乏系统性的成功概率评估模型,会导致资源错配。表现为:能看到市场动态但无法量化客户开发的成功率,或者在面对核心客户时无法清晰表达自身价值主张(Value Proposition)。这种“有感觉没体系”的状态会导致投入产出比难以预测,建议通过建立系统性的账户计划(Account Planning)来解决。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-10", "11-6"], "skills": ["xy-ai-workflow", "xy-question-spec"], "related": null, "q": 2}
{"id": "XY-CE-5965", "type": "method", "confidence": "medium", "topics": ["选品逻辑", "团队与模式设计", "AI与工具"], "knowledge": "产品开发双轨制(Global + Local):在成熟行业中,应采取“全球研发资源+本土定制能力”的组合模式。利用全球成熟的技术底座(Global R&D)确保产品的基础性能与安全性,同时建立本土定制团队(Local Customization Team),专门负责收集市场趋势、客户特殊需求并进行产品微调。在生意实战中,这意味着你不能只靠搬运总部的产品,必须针对本土客户(如特定渠道或细分人群)的特殊痛点进行“最后一公里”的适配,否则会在竞争中因缺乏灵活性而被本土品牌通过差异化功能(如更简单的操作或更好的外观)超越。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-4", "11-5"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-selection"], "related": null, "q": 2}
{"id": "XY-CE-5972", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "“基于用户调研的支付意愿定价法”:在开发新功能或配套产品时,必须通过对患者、医生、护士的多维度调研(Survey),量化用户对特定解决方案的“价值感知度”和“支付意愿(Willingness to pay)”。例如,通过调研确定患者愿意为一年的依从性方案支付约30元人民币,从而倒推产品的定价模型与生命周期价值(LTV)。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-5", "09-2"], "skills": ["xy-ai-workflow", "xy-question-spec"], "related": null, "q": 2}
{"id": "XY-CE-5997", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "技能缺口补偿机制(Skill Gap Compensation)是指当企业内部人才储备无法支撑战略实施时,通过三种路径进行补偿:1. 内部培养(如职务轮调、高管培训);2. 外部获得(通过严谨的筛选程序引进顶尖人才);3. 外部资源利用(包括外包、租赁或结盟)。在生意规模扩张期,若过度依赖内部培养而忽视外部引入,可能导致组织更新能力不足。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "12-2"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-6542", "type": "insight", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "双重角色运作模式:在构建新商业生态或平台时,运营者应具备“未来探索者”与“实践布道者”的双重身份。前者负责前瞻性技术(如AI、IoT)的用例探索,验证未来5-10年的可能性;后者负责将成熟的技术转化为可复制的、能带来真金白银的应用案例,通过“生产现场展示中心”或“培训中心”的形式进行宣贯。这种模式能有效解决新业务在推广初期的信任问题,实现从“看得到”到“用得好”的转化。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-8", "05-2"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-6564", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "技能管理智能化(Skill Management)通过追踪员工的学习曲线与实时技能水平,实现生产线的“实时线平衡”。在生意场景中,通过数字化系统匹配高技能员工与复杂工序,并针对技能瓶颈进行预警,可有效提升小批量生产下的产线整体效率。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "01-7"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-6567", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "数字化技能管理(Skill Database)通过建立基于大数据的员工技能数据库,实现从“人找活”到“系统配人”的转变。通过对员工缝纫数据、质检反馈及考试结果进行数字化记录,系统能自动识别模板开发需求并生成针对性的培训计划。在生意场景中,这解决了制造企业高度依赖熟练工的痛点,通过降低技能门槛和自动化培训路径,实现人员的高效调配与规模化扩张。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "12-3"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-6570", "type": "insight", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "透明化质量管理(Transparent Quality Management)认为高质量的数字化追溯是降低成本和提高销量的基础。通过引入视觉技术(AI识别)与IoT参数监控,将质量控制点从“事后检验”前移至“事前预防”。在生意场景中,通过对面料参数、缝纫张力等关键指标的实时监控,可以显著降低因人工操作不当导致的报废率(如目标将水洗报废率降至2%)。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-2", "12-3"], "skills": ["xy-ai-workflow", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-6644", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "员工技能矩阵(Skill Matrix)管理法:通过对各项业务技能定义难度等级,并结合近期效率达标率、质量达标率进行综合评分。将员工分为不同等级(如普通工、熟练工、多能工),作为薪酬奖惩和定制化培训计划的依据。在生意场景中,这能解决“人岗不匹配”导致的交付质量波动,通过识别员工技能缺口(Gap)来精准分配订单和培训资源。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-7", "11-6"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CE-7416", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "技术迭代的风险不对称性:当一项新技术(如AI)具备改变文明格局的潜力时,其带来的风险往往具有“不可逆性”和“滞后监管性”。由于法律法规的制定通常滞后于技术引发的社会伤亡,对于高风险、高爆发力的技术,必须在事故发生前建立预警机制。在生意场景中,这提醒操盘者:对于具有颠覆性的新增长引擎(如AI驱动的自动化运营),在规模化扩张前,必须先建立底层逻辑的“安全边界”,否则一旦算法或模型出现偏差(如价值观偏离事实),其造成的品牌声誉损失是无法通过事后公关修复的。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "06-8"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CF-0810", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "行业技能标准共建机制(Industry Skill Guild Model):通过整合产业链上下游的头部企业(如某快餐巨头、某航空集团)与行业协会,共同制定统一的职业技能标准与认证体系。在生意场景中,当你的业务进入高壁垒、重服务的领域时(如高端定制、专业咨询),不能仅靠个体员工的自觉,而应通过建立行业或细分领域的“技能地图”和“认证等级”(如金牌/银牌标准),将服务质量标准化,从而提升整个行业的溢价能力和人才留存率。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["09-1", "11-3"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-playbook"], "related": null, "q": 2}
{"id": "XY-CF-1697", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "三圈联动运营模型:构建卓越运营的底层框架,由理念与能力(MC)、机制体制(MI)和技术系统(OS)三个环环相扣的圈层组成。理念提供方向与人才支撑,机制通过KPI-KAI分解和可视化管理确保落地,技术系统提供工具与流程支撑。在生意场景中,这意味着不仅要定KPI(结果指标),还要拆解KAI(关键行动指标),并配合标准化的SOP和复盘机制,确保战略不只是口号。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "11-7"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CF-1714", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "战略管控模式(Strategic Control):适用于业务结构有一定相似性、业务间有明显协同效应的情况。总部通过战略引领(Shaping)、服务(Servicing)、监管(Safeguarding)和能力建设(Skill)四大职能发挥作用。判断标准:当企业追求多元化扩张且希望总部成为价值创造主体时,应采用此模式。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-playbook"], "related": null, "q": 2}
{"id": "XY-CF-2062", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "生成式AI(Gen AI)的供需错配与规模化红利。当前市场呈现出极高的技能溢价,尤其是生成式AI人才的供给与需求比(Talent to Demand Ratio)仅为0.3x,这意味着具备Gen AI能力的复合型人才在市场上极度稀缺。对于生意操盘手而言,这意味着:1. 降本增效的切入点在于将AI从“实验室工具”转向“业务流程集成”,例如通过定制化机器人解决传统客服无法覆盖的剩余需求;2. 竞争壁垒在于“本地化深度”,即模型能否理解特定行业的黑话、俚语及业务逻辑,而非仅仅是通用对话。在私域或营销场景中,能够利用AI生成高度贴合特定人群(如案例中的女性足球运动员)情感共鸣的内容,是实现规模化个性化的关键。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "08-3"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CF-2074", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "端到端工作流重构法:利用生成式AI(Gen AI)的开箱即用能力,结合企业私有数据进行微调,将原本过于复杂、无法拆解的业务环节(如长文本审核、多维数据比对)整合为自动化的端到端工作流。在生意场景中,这意味着不再仅仅把AI当作聊天工具,而是要将其嵌入到业务核心链路(如保险核保、供应链决策)中。判断标准在于:该方案是否解决了从数据输入到最终业务动作(如发出预警、执行指令)的闭环问题,而非仅仅停留在“辅助生成内容”阶段。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "11-1"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": ["XY-CF-2400"], "q": 2}
{"id": "XY-CF-2077", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "工业化AI落地路径:实现AI从MVP(最小可行性产品)向企业级规模化部署转化的核心在于建立MLOps/LLMOps(机器学习/大模型运维)体系。这包括:1. 插件化集成,确保AI能无缝接入现有IT/云架构;2. 数据治理,解决数据组织与可用性问题;3. 变革管理,通过培训和最佳实践分享驱动员工习惯改变。对于生意操盘手而言,这意味着在推广新工具时,不能只看技术指标,更要关注“数据质量+流程标准化+人员培训”的组合拳。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "11-11"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CF-2083", "type": "method", "confidence": "medium", "topics": ["私域运营", "团队与模式设计", "AI与工具"], "knowledge": "工业化AI运维(MLOps)方法论:指通过标准化的软件工具链,实现从数据管理、模型开发、部署到实时监控的全生命周期自动化。在生意实战中,这意味着当你试图用AI(如智能客服、自动文案生成)来规模化你的私域运营时,不能只靠“人工调优”,而必须建立一套监控机制(如监测回复准确率、检测模型是否产生幻觉/错误信息),确保AI输出的稳定性,否则规模化带来的将是成倍的品牌公信力风险。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "08-9"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CF-2089", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "合规与风控", "AI与工具"], "knowledge": "工业化AI/ML的组织适配法:当企业试图将人工智能(尤其是生成式AI)从“实验阶段”推向“规模化应用”时,必须重新定义组织架构中的责任边界。这不仅仅是引入一个工具,而是要解决三个核心问题:1. 流程集成:如何将AI模型无缝嵌入现有的业务工作流(如从营销策划到自动发文);2. 角色重定义:明确谁对AI生成的内容或决策负责(Accountability),防止算法黑盒导致的合规风险;3. 技能重塑:如何对现有员工进行大规模再培训。判断标准是:AI是否从一个“好玩的插件”变成了业务流程中不可或缺的“标准组件”。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "06-10"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-mode", "xy-ops", "xy-playbook", "xy-publish-guard"], "related": null, "q": 2}
{"id": "XY-CF-2094", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "技术落地成熟度判断:一项新技术(如AI生成代码)能否从“小众尝试”走向“主流应用”,取决于三个核心维度的演进:1. 投资回报率(ROI)的量化,例如在复杂任务中能否实现超过25%的效率提升;2. 法律与合规框架的建立,即明确技术故障时的责任归属以消除风险顾虑;3. 适用场景的覆盖面,即工具能否满足绝大多数常规业务场景的需求。对于生意操盘手而言,在引入新生产力工具时,不应仅看功能强大与否,而应重点评估其在现有业务流程中的增量效率是否可衡量,以及潜在的合规风险成本。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-2", "11-8"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": ["XY-CF-2275"], "q": 2}
{"id": "XY-CF-2096", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "过度依赖自动化导致的信任侵蚀:在引入AI或低代码工具提升生产力时,存在一个严重的陷阱——即完全脱离人工校验的自动化测试与代码审查。虽然这类工具能大幅释放人力,但若缺乏人类专家的最终把关(尤其是针对安全性、质量标准及最新规范的校验),会导致错误率上升,最终造成用户信任度的崩塌。生意操盘手在推行“降本增效”工具时,必须明确“自动化负责效率,人类负责质量边界”的原则,防止技术红利被潜在的系统性风险抵消。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "08-9"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CF-2139", "type": "insight", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "边缘计算与云端协同演进:随着AI模型从“训练阶段”转向“大规模推理阶段”,业务重心正从中心化云端向靠近用户的边缘侧(Edge)转移。在生意场景中,这意味着当你的产品涉及实时交互、低延迟反馈(如智能零售终端、即时客服机器人、自动化工厂监控)时,单纯依赖云端算力会导致响应迟滞。操盘手应意识到:未来的核心竞争力不仅在于模型本身,更在于如何通过“云+边”的架构设计,在保证低延迟的同时实现数据的高效处理。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "11-1"], "skills": ["xy-ai-workflow", "xy-ops"], "related": ["XY-CF-2137"], "q": 2}
{"id": "XY-CF-2144", "type": "case", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "跨界技术赋能实战案例:某全球快餐巨头通过与云服务商建立多年期战略合作,利用边缘计算技术赋能其移动App、自助点餐机及店内设备。该模式通过结合云端的大规模数据处理能力与边缘侧的实时计算,实现了对设备性能的深度洞察,并显著降低了门店员工的操作复杂度。这为传统零售业转型提供了范本:即利用先进技术(如AI与边缘计算)将“重资产/高复杂度”的运营环节转化为“轻量化/智能化”的自动化流程。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["13-1", "10-4"], "skills": ["xy-ai-workflow", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-CF-2145", "type": "anti-pattern", "confidence": "medium", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "落地难的“技术与管理脱节”陷阱:在尝试引入高阶新技术(如AI、云原生或量子技术)时,如果仅关注技术参数而忽视了“本地化应用场景(Local Context)”的挖掘,以及缺乏中层管理层的深度参与(Management Buy-in),会导致技术无法转化为业务增量。在私域或零售生意中,这意味着你买了一套最先进的CRM或AI客服系统,但如果一线运营人员不会用、或者管理层不认可其对ROI的贡献,这套系统只会变成昂贵的“数字负担”。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "13-2"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CF-2223", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "技术驱动的效率杠杆法:利用前沿科技(如AI、先进连接技术、生物工程)来优化传统资源密集型业务的效率。例如,利用AI提升碳捕获系统的效能,或利用卫星技术实现对土壤/海洋健康的实时监测与验证。在生意场景中,这意味着当你的核心业务处于低毛利或高资源消耗阶段时,寻找“技术+传统行业”的交叉点(如AI+农业、传感器+环境监测),通过数字化手段实现对生产要素的精准控制,从而建立竞争壁垒。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-1", "06-10"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CF-2252", "type": "number", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "AI经济价值规模预测:根据相关研究估算,AI应用每年有望释放11万亿至18万亿美元的经济价值。对于中小生意人而言,这一数字预示着行业效率革命的必然性:67%的受访组织计划在未来三年加大AI投入。这意味着,不具备AI思维的生意人将在效率竞争中面临结构性淘汰。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["01-4"], "skills": ["xy-ai-workflow", "xy-goal-card"], "related": null, "q": 2}
{"id": "XY-CF-2275", "type": "insight", "confidence": "medium", "topics": ["认知与心态", "商业案例与实战复盘", "AI与工具"], "knowledge": "技术落地成熟度判定法:判断一项新技术(如AI工具、新营销玩法)能否从“尝鲜”走向“主流”,需观察三个核心维度:1. ROI的可量化性,例如在复杂任务中能否带来超过25%的效率提升;2. 风险边界的清晰度,即法律或规则能否明确界定出错时的责任归属;3. 场景覆盖度,即该工具是否能满足绝大多数常规业务场景的需求。在生意实战中,不要盲目跟风新技术,要看它能否解决核心生产力问题并具备可控的风险边界。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["10-3", "11-8"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": ["XY-CF-2094"], "q": 2}
{"id": "XY-CF-2276", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "降本增效的“释放容量”逻辑:利用自动化工具(如AI生成代码、低代码平台)将员工从重复性、标准化的“苦力活”中解放出来,其核心目的不是单纯裁员,而是通过释放人力容量(Capacity),让高价值人才转向更具创造性的架构设计、用户体验优化及核心业务逻辑开发。在私域或生意运营中,应通过工具化手段处理基础客服、重复发券等动作,将精力集中在产品迭代与高客单价转化上。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "09-5"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": ["XY-BK3-0672"], "q": 2}
{"id": "XY-CF-2277", "type": "anti-pattern", "confidence": "medium", "topics": ["合规与风控", "AI与工具"], "knowledge": "自动化依赖陷阱:在引入AI或自动化工具(如自动发货、自动回复、自动审核)时,若完全脱离人工复核(Human-in-the-loop),极易导致错误累积并侵蚀用户信任。特别是在处理涉及合规、安全或核心业务逻辑的环节时,过度依赖自动化而缺乏人工监控与纠偏机制,会导致系统性风险。在生意规模化扩张时,必须建立“自动化执行+人工抽检”的闭环。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["08-8", "11-3"], "skills": ["xy-ai-workflow", "xy-ops", "xy-publish-guard"], "related": null, "q": 2}
{"id": "XY-CF-2280", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "AI赋能下的技能重塑方法论:在引入生成式AI(Gen AI)或无代码工具(No-code)提升生产力时,管理者不应仅关注效率提升,而应重点评估两项核心能力:一是如何定义“高质量输出”的标准(即员工需要具备识别AI生成内容好坏的判断力),二是针对团队进行必要的技能再培训(Upskilling)。在私域或内容电商场景下,这意味着运营人员需要从单纯的“执行者”转型为“AI指令专家与质量审核员”,以应对技术迭代带来的岗位职责变化。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "06-10"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-ops"], "related": ["XY-CF-2099"], "q": 2}
{"id": "XY-CF-2294", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "ROI误判陷阱(The ROI Misperception Trap)概念:指在开发AI驱动或新技术应用时,决策者往往因为担心增加安全与合规成本会降低短期价值创造(Value Creation),从而在产品设计初期忽视了数字信任的投入。在实际生意操盘中,这表现为为了追求极致的获客成本(CAC)而过度采集用户信息或模糊隐私边界,这种“以牺牲信任换取短期增长”的行为,在监管趋严的环境下会导致严重的合规成本和品牌信誉损失。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-1", "11-3"], "skills": ["xy-ai-workflow", "xy-biz-scan"], "related": null, "q": 2}
{"id": "XY-CF-2342", "type": "insight", "confidence": "medium", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "技术落地中的“人才重塑”逻辑:当新技术(如机器人、AI)进入工作流程时,初期的负面影响往往在于对现有劳动力的冲击和公众认知的抵触。对于操盘手而言,真正的转型不是简单的“人机替代”,而是通过培训提升员工技能(Upskilling),使其从执行层转向管理/协作层,实现人机协同的新型岗位设计。在私域或服务型生意中,这意味着要将人力从低效重复劳动中释放出来,投入到更高价值的客户关系维护中。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "08-7"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops"], "related": null, "q": 2}
{"id": "XY-CF-2529", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "基于数据透明度的决策优化法:利用IoT、云端计算及AI技术收集实时数据,实现路径优化与决策精准化。在私域或供应链运营中,这代表着通过数字化手段实现“精细化管理”,例如利用地理位置和实时反馈来优化物流配送或库存周转,通过降低5%左右的资源浪费(如排放或冗余成本)来提升整体利润率,实现从“经验驱动”向“数据驱动”的经营模式转型。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "08-3"], "skills": ["xy-ai-workflow", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CS-0629", "type": "method", "confidence": "medium", "topics": ["团队与模式设计", "新人起步方法论", "AI与工具"], "knowledge": "辅导矩阵法(Skill-Will Matrix):根据员工的“技能水平”与“意愿程度”两个维度,将人才划分为四个象限并采取差异化管理。1.低技能/低意愿:需“带领”,提供直接支持与严格监督;2.低技能/高意愿:需“指导”,侧重传授技巧与建立信心;3.高技能/低意愿:需“鼓舞”,通过明确目标与激励手段解决动力问题;4.高技能/高意愿:需“委任”,给予高度自主权与挑战性任务。在私域运营中,针对新入职的导购或客服应侧重指导,而针对高产出的金牌店长则应侧重委任。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-6", "08-3"], "skills": ["xy-ai-workflow", "xy-coach", "xy-mode", "xy-newbie", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-CS-1905", "type": "method", "confidence": "medium", "topics": ["私域运营", "AI与工具"], "knowledge": "数字化客户旅程重塑法:通过“调研-痛点梳理-流程再造-工具赋能”的闭环实现。首先结合客户调研与投诉数据梳理关键痛点;其次通过数字化工具(如机器人、AI助手)实现流程自动化,例如将对公开户时间缩短至10分钟内、减少50%的手工录入;最后通过全渠道覆盖实现服务同步。其核心逻辑是将“业务流程”转化为“客户体验”,通过效率提升(如决策时间缩短)和成本降低实现业务增长。", "source_type": "adapted_consulting", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["09-1", "12-2"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-TF-005", "type": "method", "confidence": "medium", "topics": ["流量获取", "AI与工具"], "knowledge": "直播大屏动态调 ROI 的规则框架:每 3 分钟采样一次,与上一轮成功样本比较(第一轮只建基线);净成交 ROI 变动超 ±10% 连续 3 轮才调目标 ROI ±1,成交金额 ±30% 连续 3 轮同理,在线人数 +30% 连续 2 轮上调、低于 20 人持续 20 分钟下调;GPM 只做对比不参与调整;涨跌信号同时出现时不动;每次调整后 20 分钟冷却期内不再调。核心思想:连续多轮确认才动作 + 冷却期,防止对波动过度反应。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["05-6", "05-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-traffic"], "related": null, "q": 2}
{"id": "XY-TF-010", "type": "method", "confidence": "medium", "topics": ["AI与工具"], "knowledge": "AI 盯盘的汇报格式纪律:每轮至少报组别、任务名、状态、消耗、ROI、CTR、是否触发动作;没触发动作也要给具体原因(如「未过 25 元」「已合格结束」);触发动作要报旧任务数据、执行了什么、新任务名/预算/时长;遇到阻塞只汇报阻塞原因,不猜测、不绕过。让自动化「每轮都能被审计」,出问题才追得回来。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "05-6"], "skills": ["xy-ai-workflow"], "related": null, "q": 2}
{"id": "XY-IPB-0497", "type": "method", "confidence": "medium", "topics": ["流量获取", "商业案例与实战复盘", "AI与工具"], "knowledge": "对于刚需型 AI 产品,可以采用“一次性买单”而非“订阅制”的逻辑。当产品解决的是即时、刚需且具有“发现即付费”属性的问题时,直接按次或按消耗付费能更快速地实现利润获取。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["09-2", "12-1"], "skills": ["xy-ai-workflow", "xy-ip", "xy-playbook", "xy-precedent"], "related": null, "q": 2}
{"id": "XY-TF2-0001", "type": "method", "confidence": "medium", "topics": ["私域运营", "团队与模式设计", "AI与工具"], "knowledge": "“眼-脑-手”协同作业法:将AI自动化流程拆解为三个独立职能模块。1. 采集指令(眼):负责实时抓取后台原始数据,要求输入参数必须与系统字段完全一致;2. 策略文档(脑):负责逻辑分析与决策边界设定,是执行的灵魂;3. 操作指令(手):负责具体的页面点击与参数修改。在生意实战中,若要让AI高效辅助投放(如千川),必须先喂入操作指令建立“手”的动作,再喂入策略文档建立“脑”的逻辑,最后通过采集指令实现闭环。这种解耦设计能有效避免AI在执行复杂投放任务时出现逻辑混乱或越权操作。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-TF2-0004", "type": "anti-pattern", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "指令模糊化风险:在利用AI进行自动化操作时,若输入的参数名称(如数据指标名)与后台系统字段不完全一致,会导致“眼”失效。在实际业务中,运营者必须确保指令中的数据名称与千川等后台显示的名称“一模一样”,否则会导致采集失败或错误的决策逻辑。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-3", "04-5"], "skills": ["xy-ai-workflow", "xy-question-spec"], "related": null, "q": 2}
{"id": "XY-TF2-0015", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "千川全域数据复盘法:在进行抖音电商投放分析时,必须建立“整体-直播间-短视频”的三层指标对齐体系。不仅要关注“整体消耗”与“整体支付ROI”,更要拆解出“直播计划”与“推商品(短视频)计划”的专项表现。判断标准在于:当整体ROI与直播间/短视频专项ROI出现背离时,需通过“净ROI”(成交金额减去近一小时退款额除以消耗)来剔除退款波动,识别真实的生意利润率。在实际操盘中,应以“整体消耗”作为大盘基准,通过对比不同计划类型的专项ROI,来决定流量预算在直播间与短视频之间的动态分配比例。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["11-10", "09-7"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops", "xy-private-ops", "xy-selection"], "related": null, "q": 2}
{"id": "XY-TF2-0016", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "直播间实时监控指标对标法:在进行直播间自动化控制或人工复盘时,必须通过抓取并对比「净成交ROI」、「净成交金额(元)」、「GPM(元)」以及「实时在线人数」这四个核心维度的数据快照。判断标准在于:必须将当前成功的样本数据与前一个样本进行实时对比,而非仅看绝对值。在执行自动化监控脚本时,若发现页面上存在多个或零个「直播大屏」入口,必须立即停止操作以防数据污染。这套方法适用于高频直播场景下的实时策略调整,确保决策基于当前活跃直播间的真实数据流。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-4", "05-6"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops", "xy-peer-pick", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-TF2-0036", "type": "method", "confidence": "medium", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "数据稳定性校验法(Stable Snapshotting):在进行自动化数据采集或业务复盘时,由于直播间、电商后台等实时性极高的场景存在数据波动或加载延迟,直接抓取单点数值会导致决策失误。该方法要求对目标指标进行连续多次(如10次)的采样对比,只有当数值在短时间内保持一致时才判定为有效数据。应用场景:私域操盘手在复盘直播间投放ROI、日预算消耗或实时成交额时,必须通过观察数据是否“稳住”来判断当前业务状态是处于真实波动还是系统加载延迟,避免基于瞬时异常值做出错误的加码或减仓决策。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["05-6", "04-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops"], "related": null, "q": 2}
{"id": "XY-TF2-0037", "type": "method", "confidence": "medium", "topics": ["认知与心态", "AI与工具"], "knowledge": "数据稳定性校验法(Stable Reads Method):在进行自动化数据抓取或业务指标复盘时,由于前端页面加载延迟、异步请求或网络波动,直接读取的数值可能存在瞬时误差。该方法要求在获取目标数据后,连续进行多次(如代码中设定的重复3次)读取并对比。只有当连续两次获取的结果完全一致时,才判定当前数据为“稳定值”。在生意场景中,这对应于对GMV、转化率等核心指标进行复盘时,必须确保数据已完成全量统计且不再发生跳变,避免基于“加载中”或“错误值”做错误的经营决策。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-2", "05-6"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-ops"], "related": null, "q": 2}
{"id": "XY-TF2-0039", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "自动化投放辅助工具的“安全写操作”原则:在进行直播全域投放等高风险参数(如日预算、目标ROI)修改时,必须遵循“用户明确授权 -> 执行写入 -> 结果回读验证”的闭环逻辑。在生意实战中,这意味着任何涉及资金流向和核心转化指标的自动化操作,都不能仅以“指令发出”为准,必须通过系统回读确认数据已成功同步至平台后台(如返回status=verified),否则视为操作失败,以防止因网络延迟或页面逻辑导致的投放预算失控。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-5", "04-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-ops", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-TF2-0049", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "全域投放ROI监控法:在进行大规模广告投放时,必须建立基于“日预算”与“全域支付ROI目标”的双维度实时监控机制。通过自动化脚本或人工复盘,重点抓取“日预算(元)”与“全域支付ROI目标”这两个核心参数,确保投放动作在预设的成本控制范围内。在生意实战中,这意味着当广告平台的实时数据(如ROI)与你设定的目标值出现偏差时,必须能第一时间通过监控指令触发预警或调整策略。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-4", "11-10"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-TF2-0060", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "AI驱动的投放自动化执行流:在利用AI工具(如Codex)进行直播间千川投流时,必须遵循“环境配置-指令固化-参数注入-路径优化”的标准化作业程序。首先,必须确保运行环境(如特定浏览器)与工具链的高度兼容;其次,通过“文件拖入+指令保存”实现技能(Skill)的本地化固化,避免AI因错误尝试而产生错误的逻辑记忆;第三,通过“红字参数替换”将业务实战思路(如预算增减、投流策略)转化为AI可执行的指令集;最后,利用“热路径建立”机制对高频监控计划进行权重优化。在生意场景中,这意味着操盘手不能仅把AI当聊天工具,而应将其视为一套需要严格执行SOP的“数字员工”,通过标准化的指令输入来确保投放动作的精准度与执行效率。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["12-4", "12-3"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-TF2-0066", "type": "method", "confidence": "medium", "topics": ["私域运营", "商业案例与实战复盘", "AI与工具"], "knowledge": "本地推自动化调优法:利用AI Agent(如Codex)结合浏览器插件实现投流参数的闭环控制。核心逻辑是将具体的业务Skill(如预算调整、ROI修正)封装为指令,通过建立“触发条件-执行动作”的逻辑链路实现自动化。在生意场景中,这解决了人工盯盘效率低、反应慢的痛点。例如:通过设置定时任务,在“半小时消耗大于200”或“ROI低于设定阈值”时,自动触发“目标ROI降低0.1”或“增加预算200”的操作。应用此方法需确保在谷歌浏览器环境下保持登录状态,并针对高频操作建立“热路径”以提升响应速度。", "source_type": "adapted_other", "source_label": "外部方法论·内化改写", "chapter": null, "section": null, "blocks": ["04-5", "12-4"], "skills": ["xy-ai-workflow", "xy-ops", "xy-playbook", "xy-private-ops"], "related": null, "q": 2}
{"id": "XY-BKG-040", "type": "case", "confidence": "medium", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "直接承认失败和困难而不美化:真人分享经验时会直接说「太难了」「我从来没想过这么难」,不会为了维持人设强行包装成「轻松搞定」,这种不加修饰的挫败感陈述比精心包装的成功学叙事更容易让观众相信内容是真实经历,而不是编出来的案例。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-2"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-ip"], "related": null, "q": 2}
{"id": "XY-BKG-1608", "type": "method", "confidence": "medium", "topics": ["流量获取", "合规与风控", "AI与工具"], "knowledge": "存在游走在平台规则边缘的操作说法(例如用实拍+AI混合规避素材审核、通过私域获客绕开平台不希望的站外导流限制),此类做法是否合规、长期是否可持续未经验证,仅作现象记录,不构成推荐或判断依据,涉及平台规则边界的问题应自行查证。", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["03-4", "11-3"], "skills": ["xy-ai-workflow", "xy-publish-guard", "xy-traffic"], "related": null, "q": 2}
{"id": "XY-BM-002", "type": "insight", "confidence": "high", "topics": ["商业案例与实战复盘", "成交与话术", "AI与工具"], "knowledge": "多个AI Agent协作时不需要模拟人类式的沟通(开会讨论),而应该直接共享底层的工作记忆/上下文路径——这是发挥「硅基生物」效率优势的方式,人类式沟通对AI协作反而是效率损耗", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-close", "xy-peer-pick", "xy-precedent"], "related": null, "q": 1}
{"id": "XY-BM-043", "type": "insight", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "判断「这件事我能不能做」不要预设专业门槛太高不敢尝试——顶级金融创新(案例:某上市公司创始人用AI设计出历史上从未有过的融资方案,自己承认单靠自己做不出来)也可能是用AI几十分钟「扫」出来的,普通人手上的AI工具已经能力足够,不需要等到「学会/被认可/万事俱备」才敢行动", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3", "01-7"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-precedent"], "related": null, "q": 1}
{"id": "XY-BM-047", "type": "method", "confidence": "high", "topics": ["团队与模式设计", "新人起步方法论", "AI与工具"], "knowledge": "AI的「观察操作自动生成技能」(录制回放)大幅降低了把业务流程编码成系统的成本——一个人演示一次操作流程,AI就能生成可复用的技能文档,整个团队都能复用,不再需要每个新人重新被教一遍", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-4", "12-2"], "skills": ["xy-ai-workflow", "xy-mode", "xy-newbie", "xy-ops", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-056", "type": "principle", "confidence": "high", "topics": ["AI与工具"], "knowledge": "把自己热爱、擅长、市场需要这三个圈的重叠部分「产品化」,才能建立一块AI和别人都进不来的垄断领域;只出卖时间、盲目追着钱跑的人反而赚不到钱,能玩一辈子、越做越有能量的事才是真正的自我价值杠杆", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["07-6", "11-1"], "skills": ["xy-ai-workflow", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-057", "type": "principle", "confidence": "high", "topics": ["AI与工具"], "knowledge": "独处是听见真实自我判断的唯一路径——AI能代替几乎一切信息处理,但代替不了一个人在安静里跟自己对话想清楚要什么;在信息爆炸、答案唾手可得的环境里,能在孤独和噪音里不慌、做出独立判断的能力反而最稀缺", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["01-7"], "skills": ["xy-ai-workflow", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-060", "type": "definition", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "当前平台重点扶持的三类深度内容形式:①有获得感的知识观点口播(讲清楚道理让用户真正提升)②AI短片和记录真实美好小事的「人间观察」类短片③制作精良、信息密度高的中长「微纪录片」", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-9"], "skills": ["xy-ai-workflow", "xy-clip", "xy-content-scan", "xy-peer-pick", "xy-script-glue", "xy-term-crack"], "related": null, "q": 1}
{"id": "XY-BM-063", "type": "method", "confidence": "high", "topics": ["流量获取", "AI与工具"], "knowledge": "把答疑群里反复出现的问题方法论固化成AI技能(skill),让用户的AI agent直接执行解决,可以在不增加人力成本的前提下提供接近「代运营」的服务体验,这是今年AI出现后对低价引流款服务的一次质变升级", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3", "08-7"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-traffic"], "related": null, "q": 1}
{"id": "XY-BM-094", "type": "principle", "confidence": "high", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "AI冲击岗位的本质不是替换某个具体技能,而是在解构重塑整个公司的组织架构和人力结构——单纯拼命学AI技能、卷技能保住旧岗位,可能是在一条正在消失的赛道上竞争,连主动拥抱AI、用AI给团队提效的人也可能被裁", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["11-6", "12-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-121", "type": "method", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "成交与话术", "AI与工具"], "knowledge": "结合AI协作的深度内容思考SOP:先描述大部分人在这个问题上的普遍现象把人群圈住,再揭示大家关于这件事普遍存在的一个误解,接着深挖产生这个误解的原因(个体心理层面或社会文化层面),然后给出正确认知,最后提供一个初步站得住脚的攻略方案——这套流程可以稳定产出有深度的内容而不依赖灵感", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["06-10", "12-3"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-content-scan", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-123", "type": "insight", "confidence": "high", "topics": ["认知与心态", "内容创作与平台", "成交与话术", "AI与工具"], "knowledge": "按固定SOP持续和AI碰撞产出内容,输出本身就是一种输入过程——随着拍摄内容越来越多,知识储备不但不会枯竭,反而会比开始创作之前更丰富;担心灵感枯竭的焦虑往往是用静止而非发展的眼光在看待内容生产", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-4", "06-9"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-close", "xy-content-scan", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-124", "type": "principle", "confidence": "high", "topics": ["团队与模式设计", "商业案例与实战复盘", "新人起步方法论", "AI与工具"], "knowledge": "企业AI转型应分两个阶段:第一阶段是自下而上的个人提效——先让每个员工学会在真实工作里和AI Agent协作、拆任务、设定评估标准;第二阶段才是组织提效——重构业务流程、打通孤立数据。很多企业的误区是员工个人AI协作能力还没培养起来,就急着引入昂贵的FDE团队做全局组织改造,顺序反了", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["11-6", "12-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-newbie", "xy-ops", "xy-peer-pick", "xy-precedent"], "related": null, "q": 1}
{"id": "XY-BM-125", "type": "method", "confidence": "high", "topics": ["团队与模式设计", "商业案例与实战复盘", "AI与工具"], "knowledge": "判断团队是否进入需要FDE介入的组织转型阶段,关键信号不是看token消耗量这类表层指标,而是看员工有没有开始自发地用多个Agent、多套技能和工作流组合起来搭建系统,形成自己的人机协作系统去完成复杂的端到端任务", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["11-6", "12-4"], "skills": ["xy-ai-workflow", "xy-mode", "xy-ops", "xy-peer-pick", "xy-precedent"], "related": null, "q": 1}
{"id": "XY-BM-130", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "选赛道要同时判断「卷度」和「商业空间」两个维度——重体力投入型赛道(如Vlog)往往最卷但商业空间也最大,纯娱乐类赛道最不赚钱、商业空间最小;平台正在重点扶持的短片、AIGC、微综艺类内容目前还是相对蓝海", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["10-1", "01-4"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-139", "type": "insight", "confidence": "high", "topics": ["AI与工具"], "knowledge": "AI对资产价值判断是把双刃剑——一方面把很多轻资产型业务(如软件)的护城河打薄,因为生成成本被压到接近零;另一方面AI本身对能源和基础设施的巨大需求,反而让电网、变压器这类建设周期长达数年、难以快速扩容的重资产成为真正稀缺、卡脖子的环节", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-1"], "skills": ["xy-ai-workflow", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-141", "type": "principle", "confidence": "high", "topics": ["私域运营", "合规与风控", "AI与工具"], "knowledge": "FDE面临的最大挑战不是技术难关而是建立客户信任、推动组织内部变革——把某个岗位的活自动化技术上可能几周就能搞定,但让真正承担业务责任的人敢把AI用在有真实风险的场景上,往往需要数月的信任建设", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["08-7", "11-6"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-private-ops", "xy-publish-guard", "xy-question-spec"], "related": null, "q": 1}
{"id": "XY-BM-142", "type": "definition", "confidence": "high", "topics": ["AI与工具"], "knowledge": "FDE角色可以拆分为负责技术部署落地的「Delta」和负责重新定义问题、推动业务变革的「Echo」——大部分讨论FDE的人只在谈Delta(把AI装进去省人力),但更有长期价值的其实是Echo(推动整个业务流程用AI重新设计一遍),这条路径最大的阻力常来自组织内部担心被取代而不愿配合的人", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["11-1", "08-9"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-term-crack"], "related": null, "q": 1}
{"id": "XY-BM-143", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "用可视化的多智能体管理终端同时并行跑多个AI Agent任务,通过侧栏和状态提示直接看出「哪个智能体在等待我的输入」,解决同时开多个AI窗口时注意力被打断、逐个点开查看的低效问题", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-4"], "skills": ["xy-ai-workflow", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-144", "type": "method", "confidence": "high", "topics": ["成交与话术", "AI与工具"], "knowledge": "把所有交流(线上会议和线下面对面沟通)都自动转化成AI可处理的文本——线上用会议转录工具自动生成纪要和待办,线下用随身录音设备持续采集,最终统一交给AI在下游整理,创作者本人只需要专注于聊天本身", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-2", "06-9"], "skills": ["xy-ai-workflow", "xy-close", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-153", "type": "anti-pattern", "confidence": "high", "topics": ["商业案例与实战复盘", "成交与话术", "AI与工具"], "knowledge": "按「单个Agent多少钱」给企业AI转型项目报价是不负责任的定价方式——一个岗位背后不只是一份SOP,还牵涉知识库对接、历史数据、权限管理、系统集成等大量业务侧梳理工作,真正的工作量在这些前期梳理而不是装Agent本身", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3", "09-2"], "skills": ["xy-ai-workflow", "xy-close", "xy-peer-pick", "xy-precedent"], "related": null, "q": 1}
{"id": "XY-BM-154", "type": "anti-pattern", "confidence": "high", "topics": ["成交与话术", "AI与工具"], "knowledge": "定制化Agent容易陷入「一次性交付」和「业务持续迭代」之间的结构性矛盾——为了完成交付,乙方不得不把客户本来柔性、多变的业务需求抽象冻结成一套刚性流程,结果上线后大量真实场景没被覆盖,员工用几次就弃用", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["11-11"], "skills": ["xy-ai-workflow", "xy-close", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-156", "type": "method", "confidence": "high", "topics": ["合规与风控", "AI与工具"], "knowledge": "企业AI转型更务实的路径是自下而上——先培训员工普遍使用个人Agent、培养协作习惯,把长尾、低风险、高频的需求交给员工自己沉淀成技能和工作流;再从这些留存了几个月的真实使用记录里筛出反复出现的共性需求,才进入需要统一知识库、权限设计和FDE系统集成的阶段", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3", "11-6"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-publish-guard"], "related": ["XY-BM-126"], "q": 1}
{"id": "XY-BM-172", "type": "insight", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "企业AI项目普遍失败(约95%的试点算不出投资回报)不是因为模型能力不够——失败和成功的项目往往用的是同一批模型,真正的差别在于AI是「贴」在业务流程上(员工自己开对话框问问题,提效有限)还是「长」在流程里(把流程本身重做一遍,带来倍数级变化)", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-1", "11-10"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-precedent"], "related": null, "q": 1}
{"id": "XY-BM-175", "type": "method", "confidence": "high", "topics": ["AI与工具"], "knowledge": "用价值-可行性四象限图给候选AI自动化场景排优先级,价值高、规则清楚的先做;但更重要的是明确标出「价值虽高但坚决不碰」的区域(如涉及专业判断、对外承诺的环节)——知道什么不该自动化,比知道什么该自动化更重要", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-1", "09-1"], "skills": ["xy-ai-workflow", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-179", "type": "principle", "confidence": "high", "topics": ["商业案例与实战复盘", "AI与工具"], "knowledge": "AI落地机制的核心不是「有没有人在客户现场」,而是现场发现的误差信号(系统不适配、流程不合理、数据对不上)能不能被反向传回产品底座、驱动整个组织跟着升级——项目结束不该只是归档,而应该是给公司多沉淀出一批可以直接复用的能力资产", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-2"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-precedent"], "related": null, "q": 1}
{"id": "XY-BM-183", "type": "method", "confidence": "high", "topics": ["商业案例与实战复盘", "合规与风控", "AI与工具"], "knowledge": "使用AI编程工具时要区分「本地数据」(终端/客户端的对话记录和记忆文件存在本地电脑,账号被封也不会丢)和「云端数据」(网页端对话,账号出问题可能直接丢失)——提前用工具自带的导出/备份命令把云端对话同步到本地文件,能在账号出问题时无缝切换到新账号或替代工具继续工作", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3", "08-8"], "skills": ["xy-ai-workflow", "xy-peer-pick", "xy-precedent", "xy-publish-guard"], "related": null, "q": 1}
{"id": "XY-BM-184", "type": "method", "confidence": "high", "topics": ["私域运营", "内容创作与平台", "AI与工具"], "knowledge": "把一本书转化成能对话的AI智能体,核心是提取「作者的表达指纹」(语言风格、知识结构)而不是做简单的检索问答——回答时不额外编造内容,需要引用原文时用增强检索找到原文并引用,让智能体像陪伴式教练那样一步步帮用户把书里的方法论应用到自己的实际问题上", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-2", "07-6"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-peer-pick", "xy-private-ops"], "related": null, "q": 1}
{"id": "XY-BM-185", "type": "insight", "confidence": "high", "topics": ["团队与模式设计", "AI与工具"], "knowledge": "把书从「一次性消费品」变成可持续对话的AI智能体产品,能把原本卖几十块钱、一次性阅读的商业模式,转变成能为作者、出版方、读者多方持续创造收入的长期产品,但商用前提是必须拿到版权授权,个人使用则以拥有该书为前提", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-2", "10-5"], "skills": ["xy-ai-workflow", "xy-biz-scan", "xy-mode", "xy-ops", "xy-peer-pick"], "related": null, "q": 1}
{"id": "XY-BM-186", "type": "method", "confidence": "high", "topics": ["内容创作与平台", "AI与工具"], "knowledge": "AI编程工具报错导致对话卡住时,按复杂度递进有三种应对方式:先尝试回退到对话历史中的某一步重新推进;不行就关闭终端重开、用resume方式恢复对话继续;还不行就让AI把当前进展写成待办清单,开一个新对话读取待办清单继续往下做", "source_type": "benchmark_account_video", "source_label": "对标内容(脱敏)", "chapter": null, "section": null, "blocks": ["12-3"], "skills": ["xy-ai-workflow", "xy-content-scan", "xy-peer-pick"], "related": null, "q": 1}
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall