Why Feishu + AI Agent?
Feishu (Lark) is where enterprises run operations β messages, documents, databases, approvals. An AI agent connected to Feishu can read context from docs, act on messages, and send structured reports. The difference between an AI that gives advice and one that gets things done.
Messaging: Send, Reply, React
OpenClaw can send messages to users, groups, and bots; reply to specific messages; and handle incoming messages as triggers. Key rule: each Feishu App has its own open_id namespace β never mix App credentials or you will hit error 99992361.
Docs and Wiki
OpenClaw can read Feishu documents, append content, create new docs from Markdown, and navigate wiki structures. Use cases: auto-generating weekly reports, updating runbooks after incidents, building knowledge bases from agent outputs.
Bitable: AI-Native Database
Bitable is Feishu's spreadsheet/database hybrid β perfect for AI agents. OpenClaw can list records, create rows, update fields, and build data workflows without human involvement. Example: an agent monitoring competitor pricing and writing findings directly to a shared Bitable table.
Interactive Cards
Instead of plain text, OpenClaw can send interactive cards with buttons, forms, and dropdowns. When uncertain, send a card and let the user choose. This converts a potentially wrong autonomous action into a safe human-confirmed one.