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
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/plugin install colocation-pipelineWrote 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/skills/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/run-fleet)<a href="https://agentmods.dev/skills/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/run-fleet"><img src="https://agentmods.dev/badge/skills/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/run-fleet/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/skills/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/run-fleet"><img src="https://agentmods.dev/badge/skills/aws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow/run-fleet.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 | $0.00093 | $0.10867 |
| Opus 5.5 | $0.00037 | $0.04347 |
| Sonnet 5.5 | $0.00019 | $0.02173 |
| Haiku 4.5 | $0.00009 | $0.01087 |
Grade A, and why
run-fleet 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.
How it starts
The opening of the file, as written. The whole thing — 647 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run a model fleet through the co-location pipeline
You are handed one thing: a fleet file (fleets/*.yaml). Your job is to take it through
the pipeline to either a deployable placement or a well-evidenced answer that it does not fit.
No agent and no state machine owns this sequence. The only enforcement in the code is the
fleet-file loader, which refuses to load a fleet for search until profile has settled every
tenant's TP degree. Everything else — running plan at all, running the demand sweep, re-running
plan after profile — is convention. You are that convention. That is why this skill exists.
Each step also has a command of its own, and the command holds the same instructions for that step
alone: /fleet-preflight, /fleet-image, /fleet-plan, /fleet-sweep, /fleet-profile,
/fleet-search, /fleet-calibrate, /fleet-inspect, /fleet-export. This skill is what holds
them in order.
Serving the result is not part of this sequence. Deploying an export on KAI GPU fractions is the
kai-deploy skill and the /fleet-kai-deploy, /fleet-kai-status, /fleet-kai-drift,
/fleet-kai-load and /fleet-kai-teardown commands, and the user starts it. Never chain into it from here; say the export is ready for it.
The plugin's hooks refuse a handful of commands outright — a helm upgrade of the daemon under a
live fleet, a pattern-kill of the MPS processes, any command that reads the Hugging Face token,
--no-holder, an uninstall that does not say what it removes or says a namespace other than the
working one, and a profile or search whose harness image is not built from the current source.
They put to the user --reprofile, --allow-disjoint-only, --relax-targets-pct, a KAI deploy or
teardown the user did not start, and an edit to a fleet file's slo: block, tolerance_pct or
kv_bytes_per_token. After a profile or search they report whether the node is still held, that
plan must run again after a profile, and any search with no export beside it. A refusal or an
ask from a hook is the rule below being applied, not an obstacle to get past.
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 · 647 lines · 93 tokens per session scan A a799f7136cfb
run-fleet is a skill 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 93 tokens to every session and 10,867 once invoked, about $0.0004 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.
run-fleet compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| run-fleetaws-samples/amazon-eks-machine-learning-with-terraform-and-kubeflow | Skill | 132 | 4d ago | Apache-2.0 | A | 93 | Claude Code |
| tensorrt-llmNousResearch/hermes-agent | Skill | 263k | today | MIT | A | 18 | Claude Code, Codex |
| agent-platform-tuninggoogle/skills | Skill | 24k | yesterday | Apache-2.0 | A | 64 | Claude Code, Codex |
| google-cloud-solution-guided-gke-ai-migrationgoogle/skills | Skill | 24k | yesterday | Apache-2.0 | A | 157 | Claude Code, Codex |
| apigee-agent-gateway-governancegoogle/skills | Skill | 24k | yesterday | Apache-2.0 | A | 222 | Claude Code, Codex |
| agent-platform-endpoint-managementgoogle/skills | Skill | 24k | yesterday | Apache-2.0 | A | 64 | Claude Code, Codex |
| gke-inferencegoogle/skills | Skill | 24k | yesterday | Apache-2.0 | A | 119 | Claude Code, Codex |