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

Lark Mcp

  • 3.1k installs
  • 38 repo stars
  • Updated March 21, 2026
  • whatevertogo/feishuskill

lark-mcp is an agent skill that documents the official Feishu/Lark MCP tools for messaging, Bitable records, document search, and wiki access with correct request shapes and OAuth rules.

About

lark-mcp is the official Feishu/Lark MCP integration skill documenting messaging, group management, Bitable CRUD, document search, import, and wiki node access for agent-driven workflows. It enforces the mcp__lark-mcp__ tool prefix, separates path, params, and data request shapes, and explains useUAT tenant versus user identity for resource creation and private data access. OAuth is required for docx_builtin_search and wiki_v1_node_search; without it callers hit 99991663 permission errors. Common pitfalls are documented: content must be a JSON string, Bitable filter values must be arrays, group creation needs owner_id, and doc search uses search_key while wiki search uses query. Quick examples cover im_v1_message_create, im_v1_chat_create, bitable_v1_appTableRecord_create, and docx_builtin_search with ID prefix tables for ou_, oc_, bascn, tbl, doxcn, and wikcn tokens. Developers reach for lark-mcp when wiring agents to Feishu messaging, multidimensional tables, or knowledge bases instead of hand-authoring Open API payloads.

  • Documents im, chat, Bitable, document, and wiki MCP tools with YAML request examples and reference links.
  • Explains useUAT true for user-owned resources and OAuth-gated doc or wiki search to avoid 99991663 errors.
  • Lists ID prefix conventions for users, chats, Bitable apps, tables, documents, and wiki nodes.
  • Captures common traps: JSON-string content, array filter values, owner_id on chat create, and token type mismatches.
  • Provides error quick-reference for tool-not-found, permission, not-found, and invalid content failures.

Lark Mcp by the numbers

  • 3,101 all-time installs (skills.sh)
  • +3 installs in the week ending Aug 4, 2026 (Skillselion tracking)
  • Ranked #146 of 2,715 Automation & Workflows skills by installs in the Skillselion catalog
  • Security screen: MEDIUM risk (skills.sh audit)
  • Data as of Aug 4, 2026 (Skillselion catalog sync)
At a glance

lark-mcp capabilities & compatibility

Capabilities
feishu/lark messaging and group management · bitable record create and search · document search and import · wiki node lookup · oauth and useuat guidance
Use cases
orchestration · api development
From the docs

What lark-mcp says it does

搜索文档/知识库必须配置 OAuth
SKILL.md
content 必须是 JSON 字符串
SKILL.md
创建群组必须指定 owner_id,否则群主为机器人
SKILL.md
npx skills add https://github.com/whatevertogo/feishuskill --skill lark-mcp

Add your badge

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

Listed on Skillselion
Installs3.1k
repo stars38
Security audit2 / 3 scanners passed
Last updatedMarch 21, 2026
Repositorywhatevertogo/feishuskill

How do I call Feishu/Lark MCP tools for messages, Bitable, and document search without hitting permission or parameter format errors?

Connect Claude, Cursor or custom agents directly to Feishu/Lark for reading Bitable records, sending formatted messages, and automating project workflows.

Who is it for?

Teams wiring Claude, Cursor, or custom agents to Feishu messaging, multidimensional tables, or Lark document and wiki APIs.

Skip if: Skip when you are not using Feishu/Lark or need generic chat integrations without the Lark MCP server.

When should I use this skill?

User mentions Feishu, Lark, Bitable, bitable,飞书文档,飞书群, or needs MCP patterns for Lark messaging and table automation.

What you get

Correct MCP tool calls with proper path, params, data, useUAT, OAuth-aware search, and troubleshootable error handling for Feishu/Lark automation.

  • Bitable query configs
  • formatted Lark messages
  • workflow automation scripts

By the numbers

  • Documents four Bitable filter operators: is, contains, isGreater, isEmpty
  • Includes multi-condition and/or conjunction filter examples in YAML

Files

SKILL.mdMarkdownGitHub ↗

Lark MCP

⚠️ 重要提醒

搜索文档/知识库必须配置 OAuth

  • docx_builtin_search → 需要 --oauth
  • wiki_v1_node_search → 需要 --oauth

否则返回 99991663 错误。配置方法见 installation.md

---

核心规则

# 工具命名(连字符,非下划线)
✅ mcp__lark-mcp__tool_name
❌ mcp__lark_mcp__tool_name

# 参数结构
path: {app_token, table_id}   # URL路径参数
params: {page_size, ...}      # 查询参数
data: {fields, ...}           # 请求体
useUAT: false                 # true=用户身份, false=租户身份

常见陷阱

# content 必须是 JSON 字符串
❌ content: {"text": "hello"}
✅ content: '{"text": "hello"}'

# 过滤条件 value 必须是数组
❌ value: "已完成"
✅ value: ["已完成"]

# 创建群组必须指定 owner_id,否则群主为机器人
owner_id: "ou_xxxxx"

# 参数名差异
docx_builtin_search: search_key  # 不是 query
wiki_v1_node_search: query       # 不是 search_key

# token 类型
wiki_v2_space_getNode: 用 wikcn...  # 不能用 doxcn...
docx_v1_document_rawContent: 用 doxcn...

useUAT 选择

场景useUAT
创建资源(想让用户可访问)true
搜索文档/知识库true
访问用户私有数据true
查询公共数据false

工具速查

类别工具文档
消息im_v1_message_create, im_v1_message_listim.md
群组im_v1_chat_create, im_v1_chat_list, im_v1_chatMembers_getchat.md
多维表格bitable_v1_app_create, bitable_v1_appTableRecord_search/create/updatebitable.md
文档docx_builtin_search, docx_v1_document_rawContent, docx_builtin_importdocuments.md
知识库wiki_v1_node_search, wiki_v2_space_getNodewiki.md

ID 类型

前缀类型来源
ou_用户IDAPI返回
oc_群聊IDim_v1_chat_list
bascn多维表格URL中 base/
tbl数据表URL参数 table=
doxcn文档搜索结果或URL
wikcn知识库节点知识库URL

快速示例

# 发送消息
工具: mcp__lark-mcp__im_v1_message_create
data:
  receive_id: "oc_xxxxx"
  msg_type: "text"
  content: '{"text": "消息内容"}'
params:
  receive_id_type: "chat_id"

# 创建群组
工具: mcp__lark-mcp__im_v1_chat_create
data:
  name: "群名"
  chat_mode: "group"
  owner_id: "ou_xxxxx"
  user_id_list: ["ou_xxxxx"]
params:
  user_id_type: "open_id"

# 创建多维表格记录
工具: mcp__lark-mcp__bitable_v1_appTableRecord_create
path:
  app_token: "bascnxxxxxx"
  table_id: "tblxxxxxx"
data:
  fields:
    文本字段: "值"
    单选字段: "选项名"
useUAT: true

# 搜索文档
工具: mcp__lark-mcp__docx_builtin_search
data:
  search_key: "关键词"
  count: 10
useUAT: true

错误速查

错误原因解决
tool not found服务器名错误使用 mcp__lark-mcp__ 前缀
99991663权限不足useUAT: true 或配置 OAuth
131005 not foundtoken 类型错误检查用 wikcn 还是 doxcn
创建资源无法访问租户身份创建使用 useUAT: true
field not found字段名错误appTableField_list 确认
invalid content格式错误content 用单引号包裹 JSON

详细文档: troubleshooting.md | installation.md

Related skills

FAQ

Why does document search return error 99991663?

docx_builtin_search and wiki_v1_node_search require OAuth configured with --oauth; without it the API returns permission error 99991663.

When should useUAT be true?

Set useUAT true when creating resources users must access, searching documents or wikis, or reading private user data; use false for public tenant queries.

How must message content be formatted?

The content field must be a JSON string such as '{"text": "hello"}', not a raw JSON object.

Is Lark Mcp safe to install?

skills.sh reports 2 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.

Automation & Workflowsagentsautomation

This week in AI coding

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

unsubscribe anytime.