AI agents use create_hook to create or update resources in Zhook — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Zhook environment.
This tool creates new webhooks/MQTT hooks, which are configurations that enable external systems to send data to the user's infrastructure. While it doesn't directly access or delete data, it does persistently create new integrations. The Write category fits because the operation creates new data/configuration entries that can be modified or removed later.
From the tool's definition The tool name is 'create_hook' and the description states it will 'Create a new webhook or MQTT hook' and return 'the new hook ID and its public URL'. This is clearly a creation operation that modifies data (adds a new webhook configuration) reversibly.
Attacks that exploit this kind of access
Create a new webhook or MQTT hook. Returns the new hook ID and its public URL. It is categorised as a Write tool in the Zhook MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Zhook MCP server in PolicyLayer and add a rule for create_hook: 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 Zhook. Nothing to install.
create_hook 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 create_hook 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 create_hook. 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.
create_hook is provided by the Zhook MCP server (zhookteam/zhook-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 →