bilibili-summarizer
Skill9 starsChanged 10 months agoMITView on GitHub
B站视频内容总结,提取核心要点和时间戳
In this repository15 other files it gives its agents.
- article-image-generatorSkill
- code-reviewer-cnSkill
- douyin-scriptSkill
- douyin-trendingSkill
- ecommerce-copywriterSkill
- email-writer-cnSkill
- knowledge-comicSkill
- meeting-notes-cnSkill
- resume-optimizer-cnSkill
- systematic-debuggingSkill
- test-driven-developmentSkill
- wechat-articleSkill
- weibo-hotSkill
- xiaohongshu-trendingSkill
- xiaohongshu-writerSkill
The file
Files
B站视频总结器
任务
分析B站视频内容,生成结构化总结,帮助快速了解视频价值。
输入
- B站视频链接 或 字幕/转录文本
- 总结风格:详细 / 精简 / 学习笔记
- 用户背景(可选):新手 / 进阶 / 专家
输出格式
# 📺 视频总结
## 基本信息
- **标题**:[视频标题]
- **UP主**:[UP主名称]
- **时长**:[HH:MM:SS]
- **分区**:[视频分区]
---
## ⭐ 推荐指数:X/5
## 📊 内容密度:X%
[干货内容占比说明]
## 🎯 适合人群
- 适合:[人群描述]
- 不适合:[人群描述]
---
## 📌 一句话总结
> [用一句话概括视频核心内容]
---
## 💡 核心要点
### 1. [要点标题] [时间戳]
[要点内容]
### 2. [要点标题] [时间戳]
[要点内容]
### 3. [要点标题] [时间戳]
[要点内容]
---
## ⏱️ 时间线
| 时间 | 内容 |
|------|------|
| 00:00 | [开场/引入] |
| XX:XX | [重点1] |
| XX:XX | [重点2] |
| XX:XX | [总结] |
---
## 📝 详细笔记
[如果选择详细模式,展开每个要点]
---
## 🎴 复习卡片
**Q1**: [问题]
**A1**: [答案]
**Q2**: [问题]
**A2**: [答案]
---
## 🔗 相关推荐
- [相关视频/资源建议]
---
## 🎬 观看建议
- **完整观看**:[是否值得完整看]
- **重点片段**:[建议重点看的时间段]
- **可跳过**:[可以跳过的部分]评估标准
推荐指数
- 5星:内容密度高,讲解清晰,强烈推荐
- 4星:质量不错,值得观看
- 3星:一般,可看可不看
- 2星:内容较水,不推荐
- 1星:标题党或内容质量差
内容密度
70%:高密度,信息量大
- 40-70%:中等,有一定水分
- <40%:低密度,水分较多
示例
输入:
视频:【硬核科普】ChatGPT到底是怎么工作的?
UP主:某科技UP
时长:25:30输出:
# 📺 视频总结
## 基本信息
- **标题**:【硬核科普】ChatGPT到底是怎么工作的?
- **UP主**:某科技UP
- **时长**:25:30
- **分区**:科技·科普
## ⭐ 推荐指数:4/5
## 📊 内容密度:65%
前5分钟为背景介绍,之后内容密度较高。
## 🎯 适合人群
- 适合:对AI感兴趣的普通用户、非技术背景想了解原理的人
- 不适合:已有深度学习基础的技术人员
## 📌 一句话总结
> ChatGPT基于Transformer架构,通过海量文本预训练+人类反馈强化学习,实现了流畅的对话能力。
## 💡 核心要点
### 1. Transformer架构 [05:30]
- 注意力机制让模型能关注重要信息
- 并行处理提升训练效率
### 2. 预训练过程 [12:15]
- 使用互联网海量文本
- 预测下一个词的任务
### 3. RLHF强化学习 [18:40]
- 人类标注员评价回答质量
- 模型学习人类偏好
## 🎬 观看建议
- **重点片段**:05:30-22:00
- **可跳过**:00:00-05:00(背景介绍)Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

