PluginBench
Skill

lark-note

open.feishu.cn

Query Feishu meeting notes by note_id to get details, type, document tokens, and unified transcripts.

What is lark-note?

Retrieves Feishu meeting note (Note) metadata and content when you have a note_id or vc-node-id. Use this to fetch note details, determine display type, access associated document tokens, and read unified verbatim records. Does not handle meeting/calendar lookup, document title search, or Docx content reading.

  • Query note metadata and document tokens (note_doc_token, verbatim_doc_token, shared_doc_tokens) by note_id
  • Determine note display type (normal, unified, or unknown) to route to correct transcript source
  • Fetch unified note transcripts with original verbatim records and speaker information
  • Route to appropriate downstream skills based on note type and user requirements

How to install lark-note

npx skills add null --skill lark-note
Prerequisites
  • lark-cli installed and configured
  • Valid note_id (from vc +detail, docs +fetch vc-node-id, or direct user input)
  • Read lark-vc/references/vc-domain-boundaries.md to understand meeting product relationships and responsibility boundaries
Claude Code
Cursor
Windsurf
Cline

How to use lark-note

  1. 1.Obtain a note_id (from user input, vc +detail, or docs +fetch vc-node-id)
  2. 2.Run `note +detail --note-id <note_id>` to retrieve note metadata and document tokens
  3. 3.Check the returned note_display_type to determine next step: if 'normal' with verbatim_doc_token use docs +fetch; if 'unified' use note +transcript; if 'unknown' handle as standalone document
  4. 4.For unified notes, run `note +transcript --note-id <note_id>` to fetch original verbatim records
  5. 5.Use returned document tokens with docs +fetch or drive metas batch_query for content or metadata

Use cases

Good for
  • User has a note_id and needs to retrieve the AI-generated summary document and verbatim transcript
  • Extract note_id from a document's vc-node-id attribute and fetch associated meeting notes
  • Determine whether a note uses unified or normal display type to decide how to access transcripts
  • Retrieve shared documents from a meeting by querying the note's shared_doc_tokens
Who it's for
  • Developers integrating Feishu meeting notes into workflows
  • Users needing to access meeting transcripts and summaries programmatically
  • Teams managing meeting documentation and follow-up tasks

lark-note FAQ

What is the difference between note_doc_token and verbatim_doc_token?

note_doc_token is the AI-generated meeting summary document (MainDoc). verbatim_doc_token is the word-for-word transcript with speaker names and timestamps (VerbatimDoc). Unified notes use note +transcript instead of verbatim_doc_token.

When should I use note +transcript vs docs +fetch?

Use note +transcript for unified display type notes to get original verbatim records. Use docs +fetch with verbatim_doc_token for normal display type notes. Check note_display_type from note +detail to determine which to use.

Can I search for notes by title or meeting ID?

No. This skill only accepts explicit note_id. For meeting ID lookup use lark-vc (vc +detail). For title search use lark-drive or lark-doc. Once you have note_id, this skill can retrieve the note details.

What if I only have a minute_token (recording)?

Use lark-minutes (minutes +detail) to extract note_id from the top level, then use that note_id with this skill. Do not use minute_token as note_id.

What does note_display_type 'unknown' mean?

Unknown type indicates the note structure cannot be automatically determined. Treat it as a standalone document and use docs +fetch with any available verbatim_doc_token. Do not assume it is unified.

Full instructions (SKILL.md)

Source of truth, from open.feishu.cn.


name: lark-note version: 1.0.0 description: "飞书会议纪要(Note)直查:已知 note_id 时查询纪要详情、展示类型、关联文档 token,并读取 unified 原始逐字记录。当用户已持有 note_id,或从文档显式 vc-node-id 获得 note_id 时使用。不负责会议/日程/妙记定位、文档标题搜索或 Docx 正文读取。" metadata: requires: bins: ["lark-cli"] cliHelp: "lark-cli note --help"

note (v1)

身份:仅使用 --as user。使用前阅读 ../lark-shared/SKILL.md

CRITICAL — 开始前 MUST 先用 Read 工具读取 ../lark-vc/references/vc-domain-boundaries.md,不读将导致命令使用、会议产物决策、领域边界职责判断错误:

  1. 了解日历 & VC、会议产物 & 文档的关联关系和职责划分
  2. 了解会议产物(妙记和纪要)之间的关联关系,例如:妙记和纪要产生条件相互独立
  3. 了解不同会议产物的组成部分,以便根据需求决策使用哪种产物的数据

Note 域只接受显式 note_id:用户直接提供,或 docs +fetch 返回的 <vc-transcribe-tab vc-node-id="..."> 中的 vc-node-id。不要从 doc_token、标题、正文或 backlink 反推 note_id

命令路由

用户表达 / 上下文路由
已知 note_id,查纪要类型 / 文档 tokennote +detail --note-id NOTE_ID
docs +fetch 返回 <vc-transcribe-tab vc-node-id="...">vc-node-id 作为 NOTE_ID,先 note +detail --note-id NOTE_ID
只持有 meeting_idvc +detail --meeting-ids <id>note_id,再 note +detail --note-id NOTE_ID
只持有 minute_token(妙记 URL)minutes +detail --minute-tokens <token> 顶层取 note_id,再 note +detail --note-id NOTE_ID(不要把 minute_tokennote_id
只持有日程 event_idcalendar +meeting --event-ids <id>meeting_id,再按上一行继续
已知 note_id,读纪要正文note +detaildocs +fetch --doc <note_doc_token>
已知 note_id,查 unified 原始记录 / 逐字稿note +transcript --note-id NOTE_ID
只有自然语言纪要标题,用户要逐字稿 / 原始记录 / 谁说了什么不进本 skill;先走文档搜索与 docs +fetch,拿到 vc-node-id 后再回来

note_display_type 路由

note +detail 结果用户要逐字稿 / 原始记录时
normal + verbatim_doc_token 非空docs +fetch --doc <verbatim_doc_token>
unknown + verbatim_doc_token 非空先按独立文档处理;不要猜成 unified
unknown + 无逐字稿 token停止重试并说明无法确定逐字稿入口
unifiednote +transcript --note-id <note_id>

判别键是 note_display_type,不是 verbatim_doc_token 是否为空:unified 纪要也可能返回非空 verbatim_doc_token

关键字段

  • note_id:Note 域唯一入口。
  • note_display_typeunknown / normal / unified
  • note_doc_token:纪要正文文档,正文读取交给 lark-doc
  • verbatim_doc_token:普通纪要逐字稿文档;unified 逐字稿不按这个 token 路由。

不在本 Skill 范围

  • 通过 meeting_id 定位纪要(note_id)→ lark-vcvc +detail)。
  • 通过 minute_token 定位纪要(note_id)→ lark-minutesminutes +detail 顶层返回 note_id)。
  • 通过日程 event_id 定位会议(meeting_id) / 用户绑定纪要(meeting_note) → lark-calendarcalendar +meeting)。
  • 自然语言纪要标题搜索 → lark-drive / lark-doc
  • Docx 正文读取 → lark-doc
  • 妙记基础信息与媒体文件 → lark-minutes

Shortcuts

Shortcut何时读 reference
+detail需要解释输出字段或根据展示类型继续路由
+transcript需要拉取 unified 原始记录或处理本地输出文件

核心概念

  • 会议纪要(Note):视频会议结束后生成的结构化文档,通过 note_id 标识。一个 Note 包含 AI 智能纪要文档、逐字稿文档和会中共享文档。
  • note_id:纪要的唯一标识符,可通过 vc +detail --meeting-ids 获取。
  • AI 智能纪要(MainDoc):AI 生成的会议总结与待办,对应 note_doc_token
  • 逐字稿(VerbatimDoc):会议的逐句发言记录,含说话人和时间戳,对应 verbatim_doc_token
  • 共享文档(SharedDoc):会中投屏共享的文档,对应 shared_doc_tokens

核心场景

1. 通过 note_id 获取纪要文档 Token

  1. 当用户已有 note_id,需要获取对应的 note_doc_tokenverbatim_doc_tokenshared_doc_tokens 时,使用 note +detail
  2. note_id 通常来自 vc +detail 的返回结果。
  3. 获取到文档 Token 后,可使用 docs +fetch 读取文档内容,或使用 drive metas batch_query 获取文档元信息。
# 1. 从会议获取 note_id
lark-cli vc +detail --meeting-ids <meeting_id>

# 2. 用 note_id 拿文档 Token
lark-cli note +detail --note-id <note_id>

# 3. 读取纪要文档内容
lark-cli docs +fetch --doc <note_doc_token> --doc-format markdown