Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add lichenyan0901-lgtm/viral-post-distiller --skill viral-post-distillergit clone --depth 1 https://github.com/lichenyan0901-lgtm/viral-post-distillerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/lichenyan0901-lgtm/viral-post-distiller/viral-post-distiller)<a href="https://agentmods.dev/skills/lichenyan0901-lgtm/viral-post-distiller/viral-post-distiller"><img src="https://agentmods.dev/badge/skills/lichenyan0901-lgtm/viral-post-distiller/viral-post-distiller/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/lichenyan0901-lgtm/viral-post-distiller/viral-post-distiller"><img src="https://agentmods.dev/badge/skills/lichenyan0901-lgtm/viral-post-distiller/viral-post-distiller.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00254 | $0.01757 |
| Opus 5 | $0.00127 | $0.00879 |
| Sonnet 5 | $0.00051 | $0.00351 |
| Haiku 4.5 | $0.00025 | $0.00176 |
Grade A, and why
viral-post-distiller scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
抖音爆款拆解 + 同款创作 Skill
这个 skill 在做什么
把"看一条爆款 → 自己拍脑袋分析 → 拍脑袋写同款"变成两步式产品：
第一步：丢链接 → 自动取数 → 严谨拆解出"爆款基因"
第二步：给新主题 → 基于基因直接产出 5 个标题 + 完整正文 + 标签 + 封面建议
两份产出物：
output/拆解_xxx.md— 拆解报告（看完就能复盘）output/创作_xxx.md— 同款创作（直接可发布）
严格的工作流（必须按顺序执行）
Phase 0: 环境自检
执行：python3 scripts/check_env.py
期望全绿：Python 3.9+ / requests / TIKHUB_API_KEY。任何一项失败，不要继续，把错误展示给用户。
Phase 1: 数据采集
执行：python3 scripts/fetch_post.py "<用户给的链接>"
脚本会：
- 校验链接（必须是
https://v.douyin.com/...或https://www.douyin.com/video/...） - 调 TikHub API
- 归一化数据 + 推导 viral 信号
- 缓存到
data/<aweme_id>.json
fetch 失败时不要硬编：如实告知用户错误（API Key 失效 / 余额不足 / 链接无效），让用户解决。
Phase 2: 四模块拆解
读取 references/post-anatomy-prompt.md 和 references/title-formulas.md。
按 4 个模块对作品做严谨拆解：
- 封面与标题：标题用了什么公式？埋了哪些搜索关键词？封面策略？
- 内容结构：开头怎么钩住读者？是干货/故事/测评/情绪型？信息节奏？结尾如何引导互动？
- 情绪与需求：触发了哪种核心情绪？满足了什么深层需求？目标用户画像？
- 爆款基因提炼：3-5 个关键词总结核心爆款要素 + 可复用模板 + 不可复制因素
每条结论必须有具体证据（文案/数据/标签）。
Phase 3: 输出拆解报告
按 references/output-template.md 生成 markdown，保存到：
output/拆解_抖音_<aweme_id 后 8 位>_<YYYYMMDD>.md
用 present_files 呈现给用户。
Phase 4: 主动询问用户新主题（关键交互）
报告呈现后，必须主动问用户：
✅ 拆解报告已生成。
我已经识别了这条爆款的核心基因：
- 标题公式：[本视频用的公式名]
- 核心情绪：[本视频触发的情绪]
- 内容结构：[本视频的内容类型]
要不要我用同款基因帮你写一篇新内容？
告诉我：
1. 新主题（必填）
2. 字数要求（可选，默认不限）
3. 你的账号特殊偏好（可选，比如"少用 emoji"、"风格更冷淡"）
如果暂时不需要，回复"先不用"即可。
等待用户回复。如果用户说"不用了 / 先不用 / 算了"，跳到结束； 如果用户给了新主题，进入 Phase 5。
Phase 5: 创作迁移（基于爆款基因写同款）
读取 references/creation-transfer-prompt.md。
基于 Phase 2 拆出的爆款基因 + 用户给的新主题，产出：
- 5 个标题备选（每个标注用的标题公式）
- 完整正文（严格按原爆款的段落结构、情绪节奏、长度）
- 5-8 个话题标签
- 封面建议（视觉方向描述）
按 references/creation-template.md 模板组织，保存到：
output/创作_<新主题名_截前10字>_<YYYYMMDD>.md
用 present_files 呈现给用户。
重要约束
必须做
- 每个 Phase 完成才进下一个
- 拆解必须给具体证据（文案/数据/标签），不要空话
- Phase 4 必须主动询问，不要静默等待
- 创作迁移必须按原爆款的"结构 + 情绪 + 公式"走，不要变成另一种风格
- 5 个标题备选必须用不同公式（不要 5 个全是数字型）
不准做
- 不准凭印象写"内容好""有共鸣""真诚"——这些都是空话
- 不准在数据没拿到时硬编结论
- 不准跳过 Phase 4 直接产出创作（除非用户明确说"只要拆解"）
- 不准在创作里硬塞与原爆款无关的元素（比如原爆款没用 emoji，新创作也不要乱加）
- TikHub API Key 不准写到代码里
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 140 lines · 254 tokens per session scan A 176a99343382
viral-post-distiller is a skill published in the GitHub repository lichenyan0901-lgtm/viral-post-distiller (2 stars, last pushed 4mo ago), licensed MIT. It adds 254 tokens to every session and 1,757 once invoked, about $0.0013 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
multi-wordcheck
A prohibited-word checker for content published on WeChat public accounts, Xiaohongshu and Douyin. It checks text, files, images and links against official word lists and marks matches with replacement suggestions in context.
wechat-creator
A content and account-analysis assistant for WeChat Official Accounts, China's platform for publishing articles to followers. It supports article writing, title scoring, rewrites, cover design, restricted-word checks, popular-article research, and account diagnosis.
playlet-wechat-feed
A daily news digest for popular short-drama articles on WeChat public accounts. WeChat public accounts are channels that publish articles; the digest groups topics and includes cover images, engagement data, and writing insights.
xiaohongshu-search
A search tool for popular Xiaohongshu posts, where Xiaohongshu is a Chinese social media platform. It finds content by keyword and ranks results using data-based scores.
playlet-bili-feed
A daily tracker for popular short-drama content on Bilibili, a Chinese video platform. It filters works by likes, groups them by subject and produces an HTML report with visual trend analysis.
playlet-xhs-feed
A daily news digest for popular short-drama posts on Xiaohongshu, a Chinese platform for lifestyle and product-sharing content. It filters posts by engagement, groups topics, and includes covers, interaction data, and creative insights.