Amazon EKS Machine Learning with Terraform and Kubeflow is a modular platform for running machine-learning operations on Amazon EKS, the managed Kubernetes service, including training pipelines and model-serving systems. It is intended for teams deploying distributed training and inference with tools such as Kubeflow, MLflow, Airflow, and KServe, using configurable AWS infrastructure. The catalogue entries provide commands, skills, hooks, and a plugin for working with this deployment setup.
Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflownpx agentmods add commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-loadWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-load)<a href="https://agentmods.dev/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-load"><img src="https://agentmods.dev/badge/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-load/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-load"><img src="https://agentmods.dev/badge/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-load.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00023 | $0.00485 |
| Opus 5.5 | $0.00009 | $0.00194 |
| Sonnet 5.5 | $0.00005 | $0.00097 |
| Haiku 4.5 | $0.00002 | $0.00049 |
Grade A, and why
fleet-kai-load scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Only when the user asked for a load test: it puts a harness Job on the CPU pool and keeps the
deployment's GPU node busy for the scenario's hours. From the repository's tutorials/multi-model-gpu-sharing directory, in
one command so the image is the one just ensured:
IMG=$(bash plugins/colocation-pipeline/scripts/harness-image.sh --ensure) && \
COLOCATION_HARNESS_IMAGE="$IMG" PYTHONPATH=src python3 -m colocation.kai load $ARGUMENTS
The scenario is a file under load-tests/examples/<instance type>/<fleet>/: as-fleet.yaml runs every
tenant at its fleet demand, demand-drift.yaml moves a few tenants away from it. It must be for
the fleet the deployment was exported from. A node of a split (<name>-n<i>) was exported from its
group's fleet file, so a parent fleet's scenario is refused on it for now; say so rather than
writing a scenario to get past it.
Exit 2 is a refusal before anything was applied: another fleet's scenario, a tenant the fleet
does not have, a request past a tenant's max_model_len, no harness image, or an engine not Ready
(verify <name> --wait-ready 900 first). Report it; do not edit the scenario to get past it.
On 0 it has applied the Job and returned. Report the per-tenant traffic it printed and the Job
name. The Job calibrates, then drives from the next whole clock hour. After each hour its log
(kubectl logs -n kubeflow-user-example-com job/<job>) has one SOAK HOUR line per tenant, the
client's figures. Once the hours have passed, plus five minutes, /fleet-kai-drift <name> records
what the deployment saw. Report the two side by side, and leave what to do about them to the user.
teardown of the deployment also deletes its load Jobs.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 32 lines · 23 tokens per session scan A b2d131fde8e7
fleet-kai-load is a command published in the GitHub repository aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow (132 stars, last pushed 4d ago), licensed Apache-2.0. It adds 23 tokens to every session and 485 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-09.
fleet-kai-load compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| fleet-kai-loadaws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow | Command | 132 | 4d ago | Apache-2.0 | A | 23 | Claude Code |
| cost-optimizewshobson/agents | Command | 42k | 5d ago | MIT | A | 26 | Claude Code |
| deploybutterbase-ai/butterbase-skills | Command | 536 | 14d ago | MIT | A | 20 | Claude Code |
| envvercel/vercel-plugin | Command | 336 | yesterday | none | A | 35 | Claude Code |
| firebase-publishPierry/harness-kit | Command | 4 | 1mo ago | MIT | A | 20 | Claude Code |
| scanprajapatimehul/claude-aws-cost-saver | Command | 24 | 4mo ago | MIT | A | 6 | Claude Code |
| migrationdatabayt/hogwarts | Command | 18 | 3d ago | none | A | 4 | Claude Code |