Lark Note
- 170k installs
- 16.1k repo stars
- Updated August 5, 2026
- larksuite/cli
This is a copy of lark-note by open.feishu.cn - installs and ranking accrue to the original listing.
lark-note is a Claude Code skill that queries Lark Feishu meeting notes by note_id and returns structured details, display type, linked document tokens, and verbatim unified transcripts for developers who already have a
About
lark-note is a v1.0.0 agent skill for the larksuite/cli repository that requires the lark-cli binary and routes only explicit note_id values to Lark Note APIs as an authenticated user. Developers use note +detail --note-id to fetch meeting summary metadata and linked Docx tokens, and can read unified raw transcripts when a vc-node-id is known from docs +fetch --api-version v2. The skill deliberately does not infer note_id from doc_token, titles, backlinks, or calendar search, which keeps agent calls deterministic. Reach for lark-note when an agent must pull Feishu meeting minutes or transcript text after the user or a prior docs fetch supplies a valid note_id.
- Queries note details and associated document token using explicit note_id
- Routes to unified transcript with note +transcript command
- Handles note_display_type logic: normal, unknown, unified
- Integrates with docs +fetch for verbatim_doc_token follow-up
- Strict note_id only policy – never guesses from titles or backlinks
Lark Note by the numbers
- 169,897 all-time installs (skills.sh)
- +9,243 installs in the week ending Aug 4, 2026 (Skillselion tracking)
- Security screen: MEDIUM risk (skills.sh audit)
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/larksuite/cli --skill lark-noteAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 170k |
|---|---|
| repo stars | ★ 16.1k |
| Security audit | 2 / 3 scanners passed |
| Last updated | August 5, 2026 |
| Repository | larksuite/cli ↗ |
How do you fetch Lark meeting note transcripts by note_id?
Query Lark (Feishu) meeting notes by note_id and retrieve structured details, display type, linked document tokens, and verbatim transcripts.
Who is it for?
Developers building Lark Feishu agents who already hold a note_id or vc-node-id from a prior docs fetch.
Skip if: Developers who need to discover meetings by title, calendar, or doc_token without an explicit note_id.
When should I use this skill?
The user provides a Lark note_id or vc-node-id and asks for meeting note details or transcript text.
What you get
Structured note detail JSON, display type, linked document tokens, and unified verbatim transcript text from Lark Note APIs.
- note detail metadata
- linked document tokens
- unified transcript text
By the numbers
- Skill version 1.0.0 in larksuite/cli
- Requires lark-cli binary per skill metadata
Files
note (v1)
身份：仅使用 --as user。使用前阅读 `../lark-shared/SKILL.md`。
Note 域只接受显式 note_id：用户直接提供，或 docs +fetch --api-version v2 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 token | note +detail --note-id NOTE_ID |
docs +fetch --api-version v2 返回 <vc-transcribe-tab vc-node-id="..."> | 取 vc-node-id 作为 NOTE_ID，先 note +detail --note-id NOTE_ID |
已知 note_id，读纪要正文 | note +detail → docs +fetch --api-version v2 --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 --api-version v2 --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 |
unified | note +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_type：unknown/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/calendar_event_id/minute_token定位纪要 → lark-vc。 - 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 |
note +detail
note +detail 只做一件事：按显式 note_id 返回纪要展示类型和相关文档 token。
lark-cli note +detail --note-id NOTE_ID --format jsonnote_id 来源
- 可以来自用户直接给出的
note_id。 - 如果入口是文档，先由 lark-doc 读取 Docx；只有
<vc-transcribe-tab vc-node-id="...">的vc-node-id可以作为note_id。 - 没有
vc-node-id时，不要从doc_token、标题、正文或 backlink 反推note_id。
输出后的路由
| detail 字段 | 后续动作 |
|---|---|
note_doc_token | 读纪要正文 / 总结 / 待办 / 章节：docs +fetch --api-version v2 --doc <note_doc_token> |
note_display_type=normal + verbatim_doc_token | 读逐字稿：docs +fetch --api-version v2 --doc <verbatim_doc_token> |
note_display_type=unknown + verbatim_doc_token | 先按普通独立逐字稿文档读取；不要猜成 unified |
note_display_type=unified | 读逐字稿 / 原始记录：转 `note +transcript` |
判别键是 note_display_type。即使 unified 纪要返回了非空 verbatim_doc_token，逐字稿仍按 unified 路由。
note +transcript
只在 note +detail 或 vc +notes 已确认 note_display_type=unified 时使用。普通纪要逐字稿是独立 Docx 文档，应回到 lark-doc 读取 verbatim_doc_token。
lark-cli note +transcript --note-id NOTE_ID行为契约
- CLI 会先校验该 Note 是否为
unified；不是 unified 时不拉取 transcript。 - CLI 内部自动翻页并拼接完整内容；任一页失败时整体报错，不保存半截 transcript。
- 默认保存到
./notes/{note_id}/unified_transcript.md；--transcript-format plain_text时保存为.txt。 - 目标文件已存在时会失败；用户明确要覆盖时才加
--overwrite。
何时不要用
| 场景 | 正确路由 |
|---|---|
| 只有纪要文档标题 | 先文档搜索，再 docs +fetch --api-version v2；有 vc-node-id 才回 Note 域 |
只有 Docx URL / doc_token | 先 docs +fetch --api-version v2；不要从 doc_token 反推 note_id |
note_display_type=normal | docs +fetch --api-version v2 --doc <verbatim_doc_token> |
note_display_type=unknown 且 verbatim_doc_token 非空 | 先按独立逐字稿文档读取 |
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7kLark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Pick lark-note when note_id is already known; use broader Lark docs or calendar skills when discovery is required first.
FAQ
What ID does lark-note require?
lark-note requires an explicit note_id supplied by the user or extracted from a vc-node-id attribute in docs +fetch --api-version v2 output. The skill does not derive note_id from doc_token, document titles, or backlinks.
Which lark-cli command fetches note details?
lark-note routes detail lookups to lark-cli note +detail --note-id NOTE_ID using user-scoped --as user authentication. The response includes meeting note type, linked document tokens, and access to unified raw transcripts.
Is Lark Note safe to install?
skills.sh reports 2 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.