Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
larksuite avatar

Lark Vc

  • 385k installs
  • 15.9k repo stars
  • Updated July 28, 2026
  • larksuite/cli

lark-vc is a Lark/Feishu CLI skill that queries and retrieves meeting records and minutes data including summaries, todos, and transcripts.

About

Query and search Lark/Feishu meeting records including summaries, todos, and transcripts. Agents can extract structured data from meeting minutes with native Lark authentication.

  • Search meeting records and retrieve meeting minutes
  • Extract summaries, todos, and transcripts from meetings
  • Structured JSON output for agent workflows

Lark Vc by the numbers

  • 385,470 all-time installs (skills.sh)
  • Security screen: LOW risk (skills.sh audit)
  • Data as of Jul 28, 2026 (Skillselion catalog sync)
At a glance

lark-vc capabilities & compatibility

Capabilities
meeting data retrieval · minutes summarization · todo extraction
Pricing
Free
npx skills add https://github.com/larksuite/cli --skill lark-vc

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs385k
repo stars15.9k
Security audit3 / 3 scanners passed
Last updatedJuly 28, 2026
Repositorylarksuite/cli

How do you fetch Lark meeting minutes from past calls?

Access and retrieve meeting records and minutes data from Lark/Feishu in agent workflows.

Who is it for?

Developers mining ended Feishu VC meetings for summaries, todos, transcripts, and attendee lists through lark-cli.

Skip if: Agents that must join live meetings or stream in-progress VC events should use lark-vc-agent instead of lark-vc.

When should I use this skill?

User asks about past meetings, meeting minutes, transcripts, action items, or who attended a completed Lark video call.

What you get

Meeting search results, AI summaries, chapter outlines, transcripts, action items, and participant snapshot records.

  • meeting summaries
  • transcripts
  • participant lists

By the numbers

  • Implements Lark vc v1 historical meeting and minutes commands
  • Published as version 1.0.0 in the larksuite/cli skill pack

Files

SKILL.mdMarkdownGitHub ↗

vc (v1)

CRITICAL — 开始前 MUST 先用 Read 工具读取 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md),其中包含认证、权限处理

核心概念

  • 视频会议(Meeting):飞书视频会议实例,通过 meeting\_id 标识。已结束的会议支持通过关键词、时间段、参会人、组织者、会议室等条件搜索(见 +search)。
  • 会议纪要(Note):视频会议结束后生成的结构化文档,包含纪要文档(包含总结、待办、章节)和逐字稿文档。
  • 妙记(Minutes):来源于飞书视频会议的录制产物或用户上传的音视频文件,支持视频/音频的转写和会议纪要,通过 minute\_token 标识。
  • 纪要文档(MainDoc):AI 智能纪要的主文档,包含 AI 生成的总结和待办,对应 note_doc_token
  • 用户会议纪要(MeetingNotes):用户主动绑定到会议的纪要文档,对应 meeting_notes。仅通过 --calendar-event-ids 路径返回。
  • 逐字稿(VerbatimDoc):会议的逐句文字记录,包含说话人和时间戳。

核心场景

1. 搜索会议记录

1. 仅支持搜索已结束的会议,对于还未开始的未来会议,需要使用 lark-calendar 技能。 2. 仅支持使用关键词、时间段、参会人、组织者、会议室等筛选条件搜索会议记录,对于不支持的筛选条件,需要提示用户。 3. 搜索结果存在多条数据时,务必注意分页数据获取,不要遗漏任何会议记录。

2. 整理会议纪要

1. 整理纪要文档时默认给出纪要文档和逐字稿链接即可,无需读取纪要文档或逐字稿内容。 2. 用户明确需要获取纪要文档中的总结、待办、章节产物时,再读取文档获取具体内容。 3. 读取智能纪要(note_doc_token)内容时,纪要文档的第一个 `<whiteboard>` 标签是封面图(AI 生成的总结可视化),应同时下载展示给用户:

# 1. 读取纪要内容
lark-cli docs +fetch --api-version v2 --doc <note_doc_token> --doc-format markdown
# 2. 从返回的 markdown 中提取第一个 <whiteboard token="xxx"/> 的 token
# 3. 下载封面图到聚合目录(和逐字稿、录像同目录,保持产物归拢)
#    并非所有纪要都有封面画板,没有 <whiteboard> 标签时跳过即可
lark-cli docs +media-download --type whiteboard --token <whiteboard_token> --output ./minutes/<minute_token>/cover
产物目录规范:同一会议的所有下载产物(录像、逐字稿、封面图等)统一放到 ./minutes/{minute_token}/ 目录下。这与 minutes +downloadvc +notes --minute-tokens 的默认落点保持一致,便于 Agent 聚合。显式路径(如封面图)需手动对齐到同一目录。
纪要相关文档 — 根据用户意图选择:
- note_doc_tokenAI 智能纪要(AI 总结 + 待办 + 章节)
- meeting_notes用户绑定的会议纪要(用户主动关联到会议的文档,仅 --calendar-event-ids 路径返回)
- verbatim_doc_token逐字稿(完整的逐句文字记录,含说话人和时间戳)— 用户说"逐字稿""完整记录""谁说了什么"时用这个
- 用户说"纪要""总结""纪要内容"时,应同时返回 note_doc_tokenmeeting_notes(如有)
- 用户意图不明确时,应展示所有文档链接让用户选择,而不是替用户决定
- 如果用户提供的是本地音视频文件并说"转纪要""转逐字稿",不要直接从 vc +notes 开始;应先用 minutes +upload 生成 minute_url,再提取 minute_token 调用 vc +notes --minute-tokens

3. 纪要文档与逐字稿链接

1. 纪要文档、逐字稿文档与关联的共享文档默认使用文档 Token 返回。 2. 仅需要获取文档名称和 URL 等基本信息时,使用 lark-cli drive metas batch_query 查询

# 学习命令使用方式
lark-cli schema drive.metas.batch_query

# 批量获取文档基本信息: 一次最多查询 10 个文档
lark-cli drive metas batch_query --data '{"request_docs": [{"doc_type": "docx", "doc_token": "<doc_token>"}], "with_url": true}'

3. 需要获取文档内容时,使用 lark-cli docs +fetch --api-version v2

# 获取文档内容
lark-cli docs +fetch --api-version v2 --doc <doc_token> --doc-format markdown

4. 查询参会人快照(读操作)

用户问"谁参加过这场会议""这个会议有哪些参会人""某某参会了吗"等参会人快照类问题时,使用 `vc meeting get --with-participants`:这是参会人服务端快照 API,不依赖 bot 身份参会,已结束会议也可查

lark-cli vc meeting get --params '{"meeting_id":"<meeting_id>","with_participants":true}'

选型判断表:

用户意图推荐命令所在 skill
参会人快照(谁参加过、何时入/离会,任意时点)vc meeting get --with-participants本 skill
已结束会议的发言内容vc +notesverbatim_doc_tokendocs +fetch --api-version v2本 skill
进行中会议的实时事件流(转写、聊天、共享、会中加入/离开)vc +meeting-events`lark-vc-agent`
Agent 真实入会 / 离会vc +meeting-join / vc +meeting-leave`lark-vc-agent`

资源关系

Meeting (视频会议)
├── Note (会议纪要)
│   ├── MainDoc (AI 智能纪要文档, note_doc_token)
│   ├── MeetingNotes (用户绑定的会议纪要文档, meeting_notes)
│   ├── VerbatimDoc (逐字稿, verbatim_doc_token)
│   └── SharedDoc (会中共享文档)
└── Minutes (妙记) ← minute_token 标识,+recording 从 meeting_id 获取
    ├── Transcript (文字记录)
    ├──

Related skills

How it compares

Use lark-vc for ended meeting artifacts; switch to lark-calendar for future schedules and lark-vc-agent for live session control.

FAQ

Does lark-vc cover live or upcoming meetings?

lark-vc covers ended historical meetings only. Query upcoming schedules with lark-calendar, and use lark-vc-agent for real-time join, leave, and in-meeting event streams.

How does lark-vc return attendee information?

lark-vc returns participant snapshots with vc meeting get --with-participants, queryable at any time including after the meeting ends, alongside AI-generated summaries and transcripts.

Is Lark Vc safe to install?

skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.