Analytics
OpenClaw / Cursor 社媒运营技能集:热点选题、多平台改写、数据复盘,让内容运营形成可持续增长闭环。
npx -y skills add yt203677-debug/ai-social-media-ops --skill analyticsAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 21 days oldThe repository was created 21 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
汇总各社媒账号内容的数据表现,做周期性复盘,沉淀「爆款规律」,产出运营建议,并把结论 回流给选题雷达让选题更准。当用户需要「复盘」「数据分析」「周报/月报」「爆款规律」 「哪条视频火了/为什么没火」「运营建议」时使用。默认用手动导出/录入的数据, 可选用浏览器只读接管已登录后台,只读不发布。
SKILL.md
4.4 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it
数据复盘(analytics)
把"发了哪些内容、数据怎么样"变成"下一步该发什么"。产出复盘报告 + 爆款规律,
并回流给 trend-topic-radar 让选题更准,形成闭环。
何时使用
- 周期性复盘(周报/月报)。
- 想知道某条内容为什么火/为什么没火。
- 需要基于数据给运营建议、调整选题方向时。
数据从哪来(三选一,默认第 1 种)
- 手动导出/录入(默认,最安全):把各平台创作者后台的数据导出或截图,填入
数据/entries.csv(模板见 templates/data-entry.csv)。截图可交给 AI 读取后转成表格。 - 浏览器只读(进阶,可选):用 OpenClaw 浏览器接管你已登录的创作者后台,只读数据页面,绝不发帖/改动。遇到登录/验证码时报"需人工处理",不猜测。
- 官方 API(若某平台有开放数据 API 再接)。
前置:读取配置
读同目录 config.yaml(无则读 config.example.yaml):review_cycle(周期)、kpi(核心指标)、
hit_multiplier(爆款判定倍数)、data_source(数据来源方式)、platforms。
产品线/人群从 ../trend-topic-radar/config.yaml 取,保持一致。
工作流
复制以下清单并逐步推进:
数据复盘 Task Progress:
- [ ] 1. 收集本周期数据(导出/录入/只读抓取)
- [ ] 2. 清洗为统一表格,按平台归一
- [ ] 3. 相对该账号基线判定:爆款/平款/哑弹
- [ ] 4. 拆解爆款共性 → 更新爆款规律库
- [ ] 5. 诊断哑弹原因
- [ ] 6. 产出复盘报告(周报/月报)
- [ ] 7. 回流:写「给选题雷达的建议」
步骤 1-2:收集与清洗
汇总本周期每条内容的字段(能拿到多少填多少,缺失标"无数据",不编造): 平台、发布日期、关联选题、标题、曝光/播放、点赞、评论、收藏、转发、涨粉、完播率或阅读完成率、 转化信号(主页访问/私信/加粉/点击)。不同平台指标名不同,按 reference.md 归一。
步骤 3:判定表现(相对基线,不用绝对数字)
对每个平台,用该账号该指标的中位数作基线:
- 爆款:核心指标 ≥
hit_multiplier× 中位数(默认 3×) - 哑弹:< 0.5 × 中位数
- 其余为平款 这样避免受账号体量影响,聚焦"相对自己"的好坏。
步骤 4:拆解爆款共性
对爆款提炼共性并写入 爆款规律库/patterns.md(累积沉淀):
选题方向、钩子类型、标题句式、形式(口播/演示/剧情/图文)、时长、发布时间、平台。
找"重复出现≥2 次"的特征作为可复用规律。
步骤 5:诊断哑弹
对哑弹给可能原因(钩子弱/选题偏/发布时间差/与人设不符/平台不匹配),给可验证的改进假设。
步骤 6:产出复盘报告
按 templates/review-report.md 生成,写入 复盘报告/YYYY-Www.md(周报)或 YYYY-MM.md(月报)。
含:数据概览、Top3/Bottom3、爆款规律、哑弹诊断、下一步建议。
步骤 7:回流选题雷达(闭环关键)
在报告末尾输出**「给选题雷达的建议」**:本周期验证有效的选题方向/钩子、应加码或规避的方向。
同步更新 爆款规律库/patterns.md。选题雷达下次产出时会参考这些规律(见 reference.md 的闭环说明)。
边界(务必遵守)
- 浏览器方式只读,绝不发帖/点赞/私信/改设置。
- 不编造数据;缺失即标"无数据"。
- 结论要能落到可执行动作,不空谈。
相关资源
- 指标定义 / 数据获取 / 爆款判定 / 闭环说明:reference.md
- 复盘报告模板:templates/review-report.md
- 数据录入模板:templates/data-entry.csv
- 配置样例:config.example.yaml
Gives 0 of the 12 instructions most analytics metrics skills give in ~1.5k tokens
Counted across 368 of the 369 authors here whose files we hold, read 2026-08-06
- read product marketing context before asking questionsin 18 of 368, across 12 files
- use lowercase with underscores for event namesin 16 of 368, across 6 files
- track events for decisions not vanity metricsin 15 of 368, across 5 files
- use object-action format for event namesin 15 of 368, across 8 files
- produce a tracking plan documentin 14 of 368, across 4 files
- Call RUBE_SEARCH_TOOLS first to get current schemasin 13 of 368, across 2 files
- establish consistent event naming conventions before implementingin 10 of 368, across 4 files
- Verify dimension and metric compatibility before reportingin 9 of 368, across 2 files
- Encrypt data at rest and in transitin 9 of 368, across 3 files
- use snake_case for event namesin 9 of 368, across 5 files
- monitor technical health during the testin 9 of 368, across 5 files
- use consistent property namesin 8 of 368, across 4 files
Said here and by no other author read
- collect and clean periodic data
- judge performance relative to the baseline median
- extract commonalities from top performers
- update the viral patterns library
- diagnose reasons for underperforming content
- output suggestions for the topic radar
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.