How to get picked for message queues by coding agents
Apache Kafka took 13% of 636 judged message queues sessions. What the numbers say a vendor in this category should do.
If you sell message queues, this page is the part of the market no dashboard shows you: what a coding agent does when a developer asks for message queues and never compares vendors.
The numbers come from 636 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code, spread across 12 realistic codebases, with every session read by a judge.
What coding agents choose for message queues
Across 636 judged sessions, Apache Kafka was chosen most often, in 13% of runs. Amazon SQS was second with 11%. In 17% of runs the agent wrote the code itself and adopted no product at all.
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Built in-house (no product adopted) | 106 | 17% |
| 2 | Apache Kafka | 83 | 13% |
| 3 | Amazon SQS | 72 | 11% |
| 4 | Service Bus | 53 | 8% |
| 5 | Amazon Kinesis | 45 | 7% |
| 6 | Laravel Queues | 43 | 7% |
| 7 | Cloud Tasks | 40 | 6% |
| 8 | Cloudflare Queues | 35 | 6% |
| 9 | Inngest | 25 | 4% |
| 10 | Vercel Workflow | 18 | 3% |
Full board, every run replayable: the message queues leaderboard.
What the shape of this category means
The leader takes only 13% of runs. This category is genuinely open and the ordering can be moved.
With the top product at 13%, message queues is decided in the moment, from what the agent reads and what it finds in the repository. Nothing is locked in, which is the best situation a vendor can be in and the one where the work pays fastest.
The order here is set by the quality of what an agent can read and by whether your product is already present in the codebase. Both are things you can change.
The agents do not agree with each other
In this category Claude Code, Codex and Cursor did not put the same product first.
Grok Build CLI (35 sessions) and Muse Code (137 sessions) join this table with fewer runs, so read their rows as indicative.
| Agent | Runs | Picked most often |
|---|---|---|
| Claude Code | 138 | Apache Kafka (22) |
| Codex | 192 | Amazon SQS (32) |
| Cursor | 134 | Apache Kafka (18) |
| Grok Build CLI | 35 | Apache Kafka (4) |
| Muse Code | 137 | Amazon SQS (20) |
That split decides where a vendor spends. In this category Codex ran a web search in 99% of its sessions and Claude Code in 12%, so what you publish reaches nearly all of Codex's message queues sessions and a minority of Claude Code's. Taking Amazon SQS's position with Codex is mostly a content problem. Taking Apache Kafka's with Claude Code leans more on the repository.
Who is asking changes the answer
Every request in this experiment was written as a specific kind of person. In this category the leader changes with the person.
| Who is asking | Runs | Picked most often |
|---|---|---|
| Vibe coder | 104 | Cloudflare Queues |
| Junior developer | 207 | Laravel Queues |
| Senior engineer | 107 | Amazon Kinesis |
| Enterprise team | 218 | Apache Kafka |
That is 4 different products winning message queues for 4 kinds of buyer, out of the same 636 sessions. Nobody here is winning message queues. They are each winning one kind of buyer.
If you sell to more than one of them, you need pages for each. See how to win the enterprise persona.
What you are really competing against
In 17% of runs the agent wrote the code itself rather than adopting a product. That is low enough that your competition is other vendors, but high enough to be worth watching.
Considered, and never chosen
Because the judge records every product an agent raised and not only the one it picked, this board also shows who kept reaching the shortlist and losing. In message queues the clearest case is Cloud Pub/Sub: on the table in 184 sessions, chosen in none.
| Product | Raised in | Chosen in |
|---|---|---|
| Cloud Pub/Sub | 184 sessions | 0 |
| NATS | 146 sessions | 0 |
| Upstash | 137 sessions | 0 |
| Trigger.dev | 108 sessions | 0 |
| Managed Kafka | 98 sessions | 0 |
Being rejected is a better position than being unknown, and a cheaper one to fix. The product is already in the agent's head and on the list. Whatever ended those 673 sessions is recorded in each transcript, one reason at a time.
What to do about it in message queues
- Treat the ordering as movable. The leader holds 13%, so agents are deliberating rather than defaulting, and the inputs they use can change the answer. This is the most winnable shape a category comes in.
- Pick which buyer you are for. The same message queues need written as a vibe coder landed on Cloudflare Queues, and written as an enterprise team landed on Apache Kafka. Those are two markets, and the enterprise one needs pages containing the constraint words: audit log, data residency, retention, single sign-on. See how to win the enterprise persona.
- Measure per agent. Claude Code put Apache Kafka first, Codex put Amazon SQS first, Cursor put Apache Kafka first, Grok Build CLI put Apache Kafka first, Muse Code put Amazon SQS first. A blended number for message queues describes a market that does not exist.
The work that applies to every category rather than to this one is written up separately: audit your documentation, write a quickstart an agent can follow, and how to measure install share.
Every message queue on this board
One page per product, with its install share, the per-agent split, and how often it was raised without being chosen.
- Do coding agents recommend Apache Kafka? — chosen in 13% of sessions
- Do coding agents recommend Amazon SQS? — chosen in 11% of sessions
- Do coding agents recommend Service Bus? — chosen in 8% of sessions
- Do coding agents recommend Amazon Kinesis? — chosen in 7% of sessions
- Do coding agents recommend Laravel Queues? — chosen in 7% of sessions
- Do coding agents recommend Cloud Tasks? — chosen in 6% of sessions
- Do coding agents recommend Cloudflare Queues? — chosen in 6% of sessions
- Do coding agents recommend Inngest? — chosen in 4% of sessions
- Do coding agents recommend Vercel Workflow? — chosen in 3% of sessions
- Do coding agents recommend Amazon EventBridge? — chosen in 2% of sessions
- Do coding agents recommend BullMQ? — chosen in 2% of sessions
- Do coding agents recommend Confluent? — chosen in 2% of sessions
- Do coding agents recommend Redpanda? — chosen in 2% of sessions
- Do coding agents recommend Amazon MSK? — chosen in 1% of sessions
- Do coding agents recommend Redis? — chosen in 1% of sessions
- Do coding agents recommend Temporal? — chosen in 1% of sessions
- Do coding agents recommend Cloud Pub/Sub? — raised in 184 sessions, chosen in none
- Do coding agents recommend NATS? — raised in 146 sessions, chosen in none
- Do coding agents recommend Upstash? — raised in 137 sessions, chosen in none
- Do coding agents recommend Trigger.dev? — raised in 108 sessions, chosen in none
- Do coding agents recommend Managed Kafka? — raised in 98 sessions, chosen in none
- Do coding agents recommend Apache Pulsar? — raised in 73 sessions, chosen in none
- Do coding agents recommend Event Hubs? — raised in 64 sessions, chosen in none
Where these numbers come from
636 judged sessions in message queues across 12 codebases, part of a published set of 15,051. Real coding agents at pinned versions, in sandboxes, inside realistic codebases, with a simulated project owner in the loop and a blind judge on every session. The full method is on one page: how we measured this.
Every message queues run can be replayed on the board.
<!-- generated by scripts/write-data-pages.mjs -->
Common questions
How many codebases is this based on?
636 judged sessions across 12 realistic codebases. A category only runs on repositories where its seam is open, so coverage differs: some categories ran on more than ten codebases and some on two.
What message queue do coding agents choose?
Across 636 judged sessions, Apache Kafka was chosen most often, in 13% of runs. Amazon SQS was second with 11%. The result changes by agent and by who is asking.
Do Claude Code and Codex pick the same message queue?
No. Claude Code picked Apache Kafka, Codex picked Amazon SQS, Cursor picked Apache Kafka. Measuring one agent tells you about part of the market only.
How often do agents build message queues themselves instead of installing something?
In 17% of runs the agent wrote the code itself rather than adopting a product.
How can a vendor improve its position here?
Make the quickstart run when pasted, state the current version on the documentation page, use one name across product, package and import, write pages for the symptoms users describe rather than only the category name, and get into the repository through templates and framework integrations.
Which message queues do agents consider but never choose?
Cloud Pub/Sub (raised in 184 sessions, chosen in none), NATS (raised in 146 sessions, chosen in none), Upstash (raised in 137 sessions, chosen in none), Trigger.dev (raised in 108 sessions, chosen in none), Managed Kafka (raised in 98 sessions, chosen in none). Being considered and not chosen is a different problem from being unknown, and it is usually fixable.
Where this comes from
Armature ran 15,051 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code inside 92 realistic codebases, and published every run. The numbers on this page come from that work.