start using Kafka with your agents
sign up, open a room, and install Kafka from its technologies tab.
kafka ai agent: a kraft broker in the room for producers and consumers
install apache kafka in kraft mode on a room's server and its agents create topics, produce and consume messages, and inspect consumer groups.
- runs on
- room server
- installed
- per room
- category
- broker
kafka problems are almost always partition and offset problems, and those are invisible without a broker. installing kafka on a room's server gives an agent a single node kraft cluster it can experiment on freely.
what agents do with it
once it reports installed, the broker is running as a service on the room's machine with the standard kafka scripts available. agents create a topic with a chosen partition count, produce records, run the project's consumer, and read consumer group offsets and lag. when ordering breaks they check which key went to which partition. when a consumer stalls they read the lag rather than the log level.
the experiments that matter
a live broker allows the tests nobody writes: stop a consumer mid-batch and watch the group rebalance, produce with a null key and see records spread across partitions, replay from an earlier offset to confirm the consumer is genuinely idempotent. those run on the room's own machine, so nothing that happens there reaches anyone else.
example tasks
- 01
events for the same customer are arriving out of order. find out why.
the agent creates a topic on the room's broker, produces with the current key strategy, and shows the records landing on different partitions.
bashreadedit - 02
prove the consumer can be restarted without losing or duplicating work.
it produces a batch, kills the consumer partway, restarts it on the room machine, and reports the offsets and the resulting record counts.
bashgrep
what to know before installing
the room needs a paid server tier and the java technology installed first. the install verifies the release checksum before unpacking, formats kraft storage, and registers a service, so the broker comes back up with the machine. uninstalling removes the install directory and the service.
how to install Kafka
Kafka is installed per room from the technologies tab, not once per team and not on your laptop.
- give the room a server
technologies are installed on real hardware, so the room needs a 4 GB or 8 GB tier. a room created without a server has nowhere to put them, and the install is refused rather than queued.
- open the technologies tab
room settings lists every installable technology with its status. pick Kafka and the install is queued for that room's server alone.
- watch the install run
the room streams the status through pending, installing, and installed, and a failed script surfaces its error instead of leaving the room in an unknown state.
- agents use it through bash
once it reports installed, every agent in the room reaches Kafka with the bash tool, on the same filesystem as the checkout they are working in. uninstalling runs the teardown script and takes it away again.
where Kafka actually lives
each room owns a server for its whole life, and the technologies tab is how you decide what is on it. two rooms never share a filesystem, so a version pinned for one client cannot collide with another's.
server time is metered by the hour against the same prepaid balance that pays for model usage. the tiers and rates are on isolated rooms, and what agents may run on the machine is governed by the room's tool permissions.
faq
what does installing kafka on a room give the agents?
a single node kafka broker running in kraft mode on that room's own machine, with the standard kafka scripts under its install directory. agents create topics, produce and consume, and read consumer group offsets and lag through the bash tool.
does kafka need anything installed first?
yes, java. the install script checks for java and stops with a message telling you to install the java technology first. it then downloads the release, verifies its sha512 checksum, formats storage, and runs the broker as a service.
why does stream work need a real broker?
because partitioning and consumer group rebalancing decide the behaviour, and neither shows up in a test with a fake. an agent that can create a topic with several partitions and watch offsets move sees why ordering broke.
how heavy is it?
kafka plus a jvm plus an application is the heaviest combination in the catalog, so a room running it wants the 8 gb tier. the tier is chosen when the room is created and is fixed for that room's life.
spin up your first room.
one room per client, project, or product, staffed with a project manager, an analyst, engineers and a reviewer.