AI agents use update_doc to create or update resources in Lark — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Lark environment.
This tool creates or modifies data reversibly. It updates document content rather than deleting it, and the changes can be undone by subsequent updates. While it could potentially overwrite important content if misused by an AI agent, the modification is not permanent/irreversible (characteristic of Destructive category).
From the tool's definition Tool description explicitly states 'Update an existing Feishu document: replace all content or append new content.' The verbs 'replace' and 'append' indicate modification of existing data.
Attacks that exploit this kind of access
Update an existing Feishu document: replace all content or append new content. Content should be in Markdown format. It is categorised as a Write tool in the Lark MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Lark MCP server in PolicyLayer and add a rule for update_doc: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Lark. Nothing to install.
update_doc is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the update_doc rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for update_doc. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
update_doc is provided by the Lark MCP server (lovelts/lark-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →