1. Conduid
  2. Content
  3. Video To Note
MCP server · Content

Video To Note

by · github.com/like-attract/video-to-note

本地优先的视频→结构化笔记服务：B站/抖音/本地视频，平台字幕+本地Whisper离线转写，LLM 生成带时间轴笔记；桌面应用 / MCP / Agent Skill 三种接入。

Unclaimed bilibilidouyinllmmcpskillvideo-notes
InstallView source
44Fair

Scored a month ago · breakdown

About Video To Note

Video To Note is an MCP server published by like-attract in the Content category: 本地优先的视频→结构化笔记服务：B站/抖音/本地视频，平台字幕+本地Whisper离线转写，LLM 生成带时间轴笔记；桌面应用 / MCP / Agent Skill 三种接入。.

The repository has 64 stars and 0 forks, with the last commit 13 days ago. Recent commit activity is the strongest signal that reported issues will get attention.

Install

From source
git clone https://github.com/like-attract/video-to-note

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Video To Note

Powered by Claude · Grounded in docs

I know everything about Video To Note. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

6 checks have not been run yet.

Releases

v1.4.5VideoToNo v1.4.5 · 4 Oct 2026VideoToNo v1.4.5 v1.4.5 更新 长视频和本地小模型更容易一次跑完**：以前中间任何一次模型请求被通道拒掉，整条任务直接失败、已经写出来的东西全丢。现在「整理某一段」失败只跳过那一段并把原文留在笔记里，归并失败停在当前层，结尾补写失败保留已写好的笔记，每种情况都留一句说明原因。同时切块前先把语音转写的碎行合并（碎行摊上时间戳开销会把几分钟的素材切成好几块，每块多押一次有损整理），整理与补写的输出额度改按材料字数给；一旦实测到某条通道在返回思考链，就自动给它…
v1.4.2VideoToNo v1.4.2 · 25 Sep 2026VideoToNo v1.4.2 v1.4.2 更新 测试连接不再把成功报成 500**：模型对连通性探测的回信如果被输出额度截成半个 emoji（未配对的代理对字符），本机后端会在响应序列化成 JSON 的那一步抛 UnicodeEncodeError。崩点在端点 return 之后，罩在端点里的异常处理抓不到，于是只回一句纯文本 Internal Server Error，前端落到兜底话术「测试请求失败（HTTP…
v1.4.1VideoToNo v1.4.1 · 16 Sep 2026VideoToNo v1.4.1 v1.4.1 更新 带截图的笔记导出后不再丢图**：笔记正文里的图片引用一直写的是 ./images/xxx.jpg，而截图落在任务目录的 frames/ 下——那个目录从来没存在过，所以只有网页预览（前端渲染时改写地址）能看到图，导出的 Markdown、导出的 HTML 和归档笔记全是坏链。现在：引用与磁盘一致；GET /api/download/ 在正文引用了真实存在的截图时返回「笔记.md + frames/」的…
v1.4.0VideoToNo v1.4.0 · 11 Sep 2026VideoToNo v1.4.0 v1.4.0 更新 长视频不再丢内容**：约 80 分钟（2.4 万字）以上的视频改为按时间顺序逐段完整记录——旧管线「逐段压缩 → 多层归并 → 一次成稿」的材料与输出上限都是与时长无关的常数，3 小时课因此丢约 2 小时。现在每段写完自动续写，仍失败就把该段原文附在笔记里并注明原因，内容不再消失；笔记带「内容概览 + 带时间范围的目录」，小标题写成「标题 [起点-终点]」（短视频同样）。80 分钟以内走的仍是原路径，行为与耗时不变…
v1.3.5VideoToNo v1.3.5 · 2 Sep 2026VideoToNo v1.3.5 v1.3.5 更新 只要字幕稿，不必先配 Key**：界面新增输出类型「仅转录字幕」——选中后笔记风格等设置自动收起，任务只读平台字幕或本地 Whisper 转写，全程不调用大模型、不需要任何 API Key。产出带时间轴的字幕稿，留在「最近任务」里（标题前带「转录」标记）随时回看、复制或下载，之后还能一键改走笔记路线，转录不会重跑 接上 MCP / Skill 之后，整理方式随你**：同一条转录出口对 agent 开放（MCP 工具…
Questions

About Video To Note

How do I install Video To Note?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Video To Note safe to use with an AI agent?

Its trust score is 44 out of 100 (fair). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Video To Note still maintained?

Yes — the latest release is v1.4.5 (4 Oct 2026), and the last commit was 13 days ago. The repository has 64 stars.