Kafka uses ZooKeeper so you need to first start a ZooKeeper server if you don't already have one.
If you do not want to install and have a separate zookeeper server, you can use the convenience script packaged with kafka to get a quick-and-dirty single-node ZooKeeper instance.
Kafka requires zookeeper and indeed the list of topics is stored there, hence the kafka-topics tool needs to connect to zookeeper too.
kafka-clients apis in the newer versions no longer talk to zookeeper directly, perhaps that's why you're under the impression a setup without zookeeper is possible. It is not, as kafka relies on it internally. For reference see:
http://kafka.apache.org/documentation.html#quickstart
Step 2:
Kafka uses ZooKeeper so you need to first start a ZooKeeper server if you don't already have one
Zookeeper is required for running Kafka. zookeeper is must. still if you want to see topic list without zookeeper then you need kafka monitoring tool such as Kafka Monitor Tool, kafka-manager etc.
The Kafka clients no longer require zookeeper but the Kafka servers do need it to operate.
You can get a list of topics with the new AdminClient API but the shell command that ship with Kafka have not yet been rewritten to use this new API.
The other way to use Kafka without Zookeeper is to use a SaaS Kafka-as-a-Service provider such as Confluent Cloud so you don’t see or operate the Kafka brokers (and the required backend Zookeeper ensemble).
For example on Confluent Cloud you would just use the following zookeeper free CLI command:
Newer versions of Kafka no longer requires ZooKeeper connection string to list topics, but can directly go via the Kafka brokers. kafka-topics.sh is provided in the bin/ folder when downloading Kafka. To list topics, do the following:
You have a stale version of the package with commands that no longer accept zookeeper but rather bootstrap-server as the connection. Confluent will then connect with Zookeeper internally.