AI agents use kafka_create_topic to create or update resources in MCP Kafka — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MCP Kafka environment.
This tool creates a new Kafka topic, which is a reversible write operation. Topics can be deleted later, so this is not destructive. The severity is medium because creating topics in a shared Kafka cluster could impact other services that depend on topic naming conventions or cluster capacity, but the effect is limited to topic creation and does not involve data deletion, financial transactions, or arbitrary code…
From the tool's definition Tool name 'kafka_create_topic' and description 'Create a new Kafka topic (requires write access)' explicitly indicate creation of a new resource.
Attacks that exploit this kind of access
Create a new Kafka topic (requires write access). It is categorised as a Write tool in the MCP Kafka MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the MCP Kafka MCP server in PolicyLayer and add a rule for kafka_create_topic: 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 MCP Kafka. Nothing to install.
kafka_create_topic 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 kafka_create_topic 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 kafka_create_topic. 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.
kafka_create_topic is provided by the MCP Kafka MCP server (williajm/mcp_kafka). 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 →