Content publish
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 content-publishAssembled 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
把一条内容(选题、口播稿、文章或视频文案)改写适配成各社媒平台版本,产出每个平台的 标题/正文/话题标签/封面建议/建议发布时间,并打包成「发布素材包」+ 人工发布清单。 当用户需要「一稿多平台」「多平台改写」「内容分发」「排期」「发布文案」「适配抖音/视频号/ 公众号/YouTube/X/Instagram/TikTok/亚马逊」时使用。只产出素材,不自动发布,需人工发布。
SKILL.md
4.2 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it
多平台分发(content-publish)
把一条内容改写成各平台版本并打包,解决"不想一个个平台重写文案"的痛点。 不自动发布——最终由人工按清单去各平台发。零封号风险。
何时使用
- 选好一条选题/写好一稿后,要分发到多个平台时。
- 已有一篇公众号文/一条视频,想改编成其他平台版本时。
前置:读取配置
- 产品线/人群/平台开关:读
../trend-topic-radar/config.yaml的products、audience、platforms(单一事实来源)。 - 发布相关设置(发布时间偏好、各平台账号名):读同目录
config.yaml(无则读config.example.yaml并提示补全)。 - 目标平台 =
platforms中为true的平台。
工作流
复制以下清单并逐步推进:
多平台分发 Task Progress:
- [ ] 1. 确认内容源(选题/稿件)与核心信息
- [ ] 2. 读取目标平台、产品线卖点、人设基调
- [ ] 3. 逐平台改写(套用 reference.md 的平台规范)
- [ ] 4. 每平台产出:标题/正文/标签/封面建议/建议发布时间/注意事项
- [ ] 5. 打包为发布素材包 + 人工发布清单,归档
- [ ] 6. 交用户人工发布(不自动发)
步骤 1:确认内容源
输入可以是:选题库/YYYY-MM-DD.md 里的某条选题编号、一段口播稿、一篇文章、或一条视频的核心信息。
先提炼出这条内容的核心信息(1-2 句)+ 绑定的产品线卖点,作为所有平台版本共享的内核。
步骤 2:确认目标与基调
读配置取目标平台清单与产品卖点。保持跨平台人设与核心信息一致,只改形态/语气/长度/标签,不改立场。
步骤 3-4:逐平台改写并产出
对每个目标平台,套用 reference.md 里对应的「平台改写规范」,产出该平台的:
- 标题/开头钩子(符合该平台标题习惯)
- 正文/文案(符合该平台形态与长度)
- 话题标签(数量与风格按平台规范)
- 封面/首图建议(一句话)
- 建议发布时间(按 reference.md 的时段建议 + config 偏好)
- 平台注意事项(该平台的雷区/合规点)
规则:不编造数据;涉及 avoid 清单的敏感点要规避;亚马逊按"商品文案"而非"内容帖"处理。
步骤 5:打包归档
按 templates/publish-package.md 生成发布素材包,写入 发布库/YYYY-MM-DD-<选题简称>.md。
素材包末尾必须含一份人工发布清单(勾选式,标注每平台发布时间与状态)。
步骤 6:人工发布(硬性节点)
明确告知用户:"以上为各平台发布素材包,请按清单自行发布。" 不得调用浏览器或任何接口自动发帖、登录、点赞。
与其他模块的衔接
- 上游:
trend-topic-radar(选题库)提供选题。 - 视频成片:口播稿可交给视频工厂
D:\job\项目-软件\项目\video(npm run make -- --file <稿> -f 9:16)出片,再回到本模块做多平台文案分发。 - 下游(未来):自动/半自动发布另做模块,本模块只到"素材包+清单"。
边界(务必遵守)
- 只产出文案素材,绝不自动发布/登录/私信。
- 各平台版本保持核心信息与人设一致,避免自相矛盾。
- 遵守各平台条款与内容合规,规避
avoid清单话题。
相关资源
- 各平台改写规范与建议发布时间:reference.md
- 发布素材包模板:templates/publish-package.md
- 发布相关配置样例:config.example.yaml
Gives 0 of the 12 instructions most social media skills give in ~1.3k tokens
Counted across 489 of the 492 authors here whose files we hold, read 2026-08-06
- build content around three to five pillarsin 24 of 489, across 12 files
- read product marketing context before asking questionsin 23 of 489, across 13 files
- respond to all comments on your postsin 21 of 489, across 9 files
- adapt tone and structure per platformin 18 of 489, across 8 files
- adapt content for each platformin 15 of 489, across 10 files
- use the output flag to specify an output directoryin 14 of 489, across 4 files
- Generate output logo images with white backgroundin 13 of 489, across 4 files
- Fix failing generation scripts directlyin 13 of 489, across 4 files
- ask user about html preview after logo generationin 12 of 489, across 3 files
- run the download script with a URLin 12 of 489, across 3 files
- implement exponential backoff for 429 responsesin 12 of 489, across 3 files
- include a single clear call to actionin 12 of 489, across 9 files
Said here and by no other author read
- extract core message and product卖点
- keep persona consistent across platforms
- apply platform norms from reference
- produce titles body tags and cover suggestions
- never fabricate data
- avoid sensitive topics
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.