Feishu Multiagent Setup

Use this skill when a user wants to connect multiple Feishu bots to OpenClaw and needs a safe setup checklist.

Public Safety Boundary

Do not ask the user to paste real App Secret, API Key, Verification Token, Encrypt Key, cookies, or login state into chat.
Do not write openclaw.json, auth-profiles.json, shell profiles, service files, or gateway config from this public skill.
Do not run setup scripts or restart Gateway from this public skill.
Do not bypass Feishu login, QR code, captcha, organization approval, or admin authorization.
If the user wants private execution, first explain what will be written and ask for explicit confirmation in their local environment.

Workflow

Ask how many bots they need and what each bot should be used for.
Generate a table with safe placeholders:
- bot name
- Feishu app display name
- event subscription URL placeholder
- required permissions/events
- local config destination
Guide the user through Feishu Open Platform pages by visible Chinese labels.
When credentials appear, tell the user to keep them in a local private config file, not in public docs or chat history.
Produce a local review checklist for their OpenClaw operator:
- each bot has one account id
- each account id has one route binding
- each route binding points to the intended agent
- each agent has an isolated workspace
- test messages are sent only after the user confirms setup is complete

Output Shape

Return a concise checklist plus a local-only config draft with placeholders. Example:

bot_name: customer_service_bot
feishu_app_id: cli_xxx_placeholder
secret_storage: local private config only
route: feishu:<account_id> -> agent:<agent_id>
status: pending user-side credential entry

Reference

Read references/setup-checklist.md when the user needs a full step-by-step handoff.