← 返回 Skills 市场
Feishu Cli Setup
作者
Cosmos Fang
· GitHub ↗
· v1.0.0
· MIT-0
100
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install feishu-cli-setup
功能描述
Step-by-step AI agent guide for installing and configuring lark-cli (飞书/Lark CLI). Designed for Claude, Manus, and OpenClaw to proactively guide users throug...
使用说明 (SKILL.md)
Feishu CLI Setup
An AI-native installation guide for lark-cli — the official Lark/Feishu CLI tool (6.7k ⭐, maintained by larksuite). Enables Claude, Manus, and OpenClaw to proactively guide users from zero to a fully authenticated lark-cli in minutes.
Why does this skill exist? Installing lark-cli involves a browser OAuth step that AI agents cannot complete on behalf of users. This skill provides the exact prompts and workflows for agents to extract authorization URLs from CLI output and present them to users at the right moment — making the otherwise tricky setup seamless.
What is lark-cli?
| Item | Detail |
|---|---|
| Repo | github.com/larksuite/cli |
| Latest | v1.0.4 |
| Commands | 200+ |
| Agent Skills | 20 built-in (Calendar, IM, Docs, Base, Sheets, Tasks, Mail, Wiki, …) |
| License | MIT |
| Install | npm install -g @larksuite/cli |
Scripts
| Script | Purpose |
|---|---|
scripts/check.js |
Detect OS, Node.js version, and lark-cli install state |
scripts/install.js |
Guide npm install -g @larksuite/cli + skills add |
scripts/config.js |
Guide lark-cli config init --new — extracts browser URL for user |
scripts/auth.js |
Guide lark-cli auth login --recommend — extracts OAuth URL for user |
scripts/verify.js |
Run lark-cli auth status and summarize available Agent Skills |
scripts/learn.js |
Show first commands to try (calendar, im, docs, tasks) |
scripts/setup.js |
Full guided pipeline — runs all steps in sequence |
Usage
# Full guided setup (recommended)
node scripts/setup.js
# Individual stages
node scripts/check.js # detect current state
node scripts/install.js # install lark-cli
node scripts/config.js # configure Feishu app credentials
node scripts/auth.js # authenticate via OAuth
node scripts/verify.js # verify and list available skills
node scripts/learn.js # first commands guide
# Language options
node scripts/setup.js --lang en
node scripts/setup.js --lang zh # default
Agent Workflow (4 Steps)
Step 1 — Install
npm install -g @larksuite/cli
npx skills add larksuite/cli -y -g
Step 2 — Configure app credentials (run in background, extract URL → send to user)
lark-cli config init --new
Step 3 — Login (run in background, extract URL → send to user)
lark-cli auth login --recommend
Step 4 — Verify
lark-cli auth status
20 Built-in Agent Skills (after setup)
| Skill | Domain |
|---|---|
lark-shared |
Auth, config, identity (auto-loaded) |
lark-calendar |
Calendar, agenda, events |
lark-im |
Messages, group chats, reactions |
lark-doc |
Documents (Markdown) |
lark-drive |
Files, uploads, downloads |
lark-sheets |
Spreadsheets |
lark-base |
Tables, records, views, dashboards |
lark-task |
Tasks, subtasks, reminders |
lark-mail |
Email (send, reply, search) |
lark-contact |
User search by name/email/phone |
lark-wiki |
Knowledge spaces & nodes |
lark-event |
WebSocket event subscriptions |
lark-vc |
Meeting records & minutes |
lark-whiteboard |
Whiteboard/chart DSL |
lark-minutes |
Meeting AI artifacts |
lark-openapi-explorer |
API documentation explorer |
lark-skill-maker |
Custom skill framework |
lark-approval |
Approval tasks & workflows |
lark-workflow-meeting-summary |
Meeting summary workflow |
lark-workflow-standup-report |
Standup report workflow |
Security Notes
lark-cli runs under your Feishu/Lark user identity. Keep these in mind:
- Do not share your
LARK_APP_ID/LARK_APP_SECRETin public repos - Use
--dry-runfor commands with side effects before executing - Do not add the bot to group chats if you want to avoid permission exposure
- Credentials are stored in the OS native keychain (not plaintext)
Version: 1.0.0 · Source: github.com/larksuite/cli · Updated: 2026-04-05
安全使用建议
This skill appears to be a legitimate installer/guide for the official lark-cli. Before running anything, verify you trust the upstream packages: check the npm package publisher for @larksuite/cli and the GitHub repo (https://github.com/larksuite/cli). Be cautious when performing global npm installs (avoid sudo if possible; follow npm permission guidance). When asked for App ID/Secret, paste them only into the local CLI prompt — do not paste secrets into public chats or repositories. Confirm that any authorization URL you receive is a legitimate feishu/larksuite domain before opening it. If you prefer less system-wide impact, you can install lark-cli locally instead of globally and avoid running npx skills add -g.
功能分析
Type: OpenClaw Skill
Name: feishu-cli-setup
Version: 1.0.0
The skill bundle provides a structured, multi-step guide for an AI agent to install and configure the official Feishu/Lark CLI (lark-cli). The included Node.js scripts (e.g., setup.js, auth.js, install.js) act as prompt generators that instruct the agent to perform environment checks, execute npm installations, and handle OAuth authentication flows. The instructions are transparent, include security warnings regarding credential handling, and encourage the use of safety flags like --dry-run. No evidence of data exfiltration, malicious payloads, or intentional prompt injection was found.
能力标签
能力评估
Purpose & Capability
Name/description (Feishu/Lark CLI install + OAuth guidance) match the provided artifacts. The package includes prompt-generator scripts that produce step-by-step instructions for installing/configuring lark-cli and loading the 20 Agent Skills; these are appropriate for the stated goal.
Instruction Scope
SKILL.md and all scripts only instruct the agent to run local CLI commands (npm install, npx, lark-cli commands), extract authorization/config URLs from CLI output, and present them to the user. They explicitly require user browser interaction for OAuth and encourage --dry-run and user confirmation. The instructions do not direct reading unrelated system files or exfiltrating secrets.
Install Mechanism
There is no registry install spec; this is an instruction-only skill with helper scripts. The instructions recommend installing from npm (npm install -g @larksuite/cli) or building from the public GitHub repo — both are standard distribution channels. Note: global npm installs (-g) and the npx skills add command will modify the system/global agent skill list; that's expected but has system-wide impact.
Credentials
The skill requests no environment variables or credentials itself. It guides the user to provide LARK_APP_ID/LARK_APP_SECRET to the lark-cli (entered by the user) and explicitly warns not to share secrets publicly. Mention of LARK_CONFIG_FILE in troubleshooting is contextual and not required by the skill.
Persistence & Privilege
always:false and disable-model-invocation:false (normal). The guide instructs installing global skills (npx skills add -g) which makes the lark-* skills available to all agents on the host — this is expected for providing Feishu features but is a meaningful system-wide change the user should be aware of.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install feishu-cli-setup - 安装完成后，直接呼叫该 Skill 的名称或使用
/feishu-cli-setup触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release. AI agent guide for installing lark-cli (飞书/Lark CLI) — covers environment check, npm install, Feishu app config (browser OAuth flow), auth login, and verification. Designed for Claude, Manus, and OpenClaw. Includes learn.js with first-steps examples for calendar, IM, docs, tasks, and contacts.
元数据
常见问题
Feishu Cli Setup 是什么？
Step-by-step AI agent guide for installing and configuring lark-cli (飞书/Lark CLI). Designed for Claude, Manus, and OpenClaw to proactively guide users throug... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 100 次。
如何安装 Feishu Cli Setup？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install feishu-cli-setup」即可一键安装，无需额外配置。
Feishu Cli Setup 是免费的吗？
是的，Feishu Cli Setup 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Feishu Cli Setup 支持哪些平台？
Feishu Cli Setup 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Feishu Cli Setup？
由 Cosmos Fang（@cosmofang）开发并维护，当前版本 v1.0.0。
推荐 Skills