Skip to content

Lark CLI Shared Rules

Teaches the agent how to use lark-cli for auth, identity, permissions and safe writes

Repository
larksuite/cli
Stars
17,928
Licence
MIT
Files
7 files

What it does

This is the shared foundation for the Lark (Feishu) command-line tool skills. Once installed, the agent knows how to log in and check auth, choose between acting as the user or as the app bot, request business-domain permissions, and read the CLI's JSON success/error format correctly. It also tells the agent to stop and ask you before any high-risk write, to show QR codes for authorization links, and to use only relative file paths.

Great for

  • Developers automating Lark/Feishu tasks with lark-cli
  • Teams setting up user vs bot identity and permission scopes for Lark
  • Anyone building other lark-* skills that depend on these shared rules

Skip it if

  • People who do not use Lark/Feishu or lark-cli, since the rules have no use without the tool
  • Unattended automation that must run high-risk writes without a human confirming, because the skill requires explicit user approval for those

Try asking

  • “Log me into lark-cli as a user and check which scopes are missing”
  • “Set up lark-cli config for a new Lark app and show me the authorization QR code”
  • “Revoke my lark-cli authorization for the docs domain”
  • “Explain why my lark-cli command returned exit code 10 and what I need to confirm”
Step-by-step guide: how to use Lark CLI Shared RulesInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 7 original files, the MIT licence and an attribution note.

    lark-shared.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip lark-shared.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip lark-shared.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Let the agent read the referenced files or run --help before running commands, as the skill instructs.
  • Review the action, risk and parameters the agent shows you at any exit-code-10 prompt before approving.
  • Keep app secrets out of the chat; the skill forbids printing them but you should still avoid pasting them.

What's inside

  • references/lark-shared-config-init.md
  • references/lark-shared-high-risk-approval.md
  • references/lark-shared-identity-and-permissions.md
  • references/lark-shared-output-contract.md
  • references/lark-shared-update-notice.md
  • references/lark-wiki-token-routing.md
  • SKILL.md

Redistributed unmodified from larksuite/cli under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.