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
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflowWrote 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-drift)<a href="https://agentmods.dev/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-drift"><img src="https://agentmods.dev/badge/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-drift/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-drift"><img src="https://agentmods.dev/badge/commands/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/fleet-kai-drift.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 | $0.00025 | $0.00352 |
| Opus 5.5 | $0.00010 | $0.00141 |
| Sonnet 5.5 | $0.00005 | $0.00070 |
| Haiku 4.5 | $0.00003 | $0.00035 |
Grade A, and why
fleet-kai-drift 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
From the repository's tutorials/multi-model-gpu-sharing directory:
PYTHONPATH=src python3 -m colocation.kai drift $ARGUMENTS
Read-only against the cluster: it asks the cluster Prometheus for every whole hour since the deploy
that is not recorded yet, appends one JSON line per tenant per hour to drift/<name>.jsonl beside
the export's directory, and pushes the file to EFS. no whole hour to record yet is not a fault.
Report, per tenant, what the printed lines say: the request rate and the prompt and output lengths
against the fleet file (vs fleet), and ITL mean and TTFT p95 against the target (vs target).
no traffic is an idle hour, not zero drift. too few requests has no p95. TTFT p95 comes from
vLLM's coarse histogram: report it as a trend, not as the harness's figure.
A node of a split is its own deployment, <name>-n<i>, and is recorded against its group's fleet
file.
It decides nothing, and neither do you: do not edit the fleet file or start a run because of it.
Say which tenants have moved and how far, and leave whether to re-run the pipeline to the user.
recorded here but not pushed to EFS exits 1 with the local file intact; report it.
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 · 28 lines · 25 tokens per session scan A e87d06fcb812
fleet-kai-drift 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 25 tokens to every session and 352 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-drift compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| fleet-kai-driftaws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow | Command | 132 | 4d ago | Apache-2.0 | A | 25 | 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 |