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.
npx skills add MichalSteczko/rhoai-agentic-skills --skill deploy-maas-modelgit clone --depth 1 https://github.com/MichalSteczko/rhoai-agentic-skillsWrote 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/michalsteczko/rhoai-agentic-skills/deploy-maas-model)<a href="https://agentmods.dev/skills/michalsteczko/rhoai-agentic-skills/deploy-maas-model"><img src="https://agentmods.dev/badge/skills/michalsteczko/rhoai-agentic-skills/deploy-maas-model/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/michalsteczko/rhoai-agentic-skills/deploy-maas-model"><img src="https://agentmods.dev/badge/skills/michalsteczko/rhoai-agentic-skills/deploy-maas-model.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 | $0.00041 | $0.04738 |
| Opus 5 | $0.00020 | $0.02369 |
| Sonnet 5 | $0.00008 | $0.00948 |
| Haiku 4.5 | $0.00004 | $0.00474 |
Grade A, and why
deploy-maas-model scanned grade A with 1 finding 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 11d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -X POST \ How it starts
The opening of the file, as written. The whole thing — 441 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deploy Model to MaaS
Add a model to Models-as-a-Service (MaaS) by creating the required governance resources and issuing an API key for gateway access.
Agent Behavior Rules (MANDATORY — never skip)
1. Always use AskUserQuestion for every decision point
Never ask the user a question in plain text. Every question — namespace, resource name, vLLM args, deploy mode, subscription owner, rate limits — MUST be presented via the AskUserQuestion tool with a menu of options. The user selects from a list; they do not type free-form answers.
2. Always show default vLLM args before asking the user
Before presenting the vLLM args question, run:
./get-default-args.sh "<resolved-model-name>"
This script reads model-configs.json and outputs one arg per line, detecting family + variant + size automatically. Display the full output as a code block in the first option of the AskUserQuestion menu, with the label:
"Use defaults (Recommended)"
Always offer at least one common override option (e.g. reduced --max-model-len) and an "Other" free-text fallback. The user must never be asked about vLLM args without seeing the defaults first.
Namespace Reference
| Resource | Namespace |
|---|---|
| MaaSSubscription, MaaSAuthPolicy | models-as-a-service |
| MaaSModelRef | model's own namespace (e.g. <model-namespace>) |
| maas-api, maas-postgres | redhat-ai-gateway-infra |
| Kuadrant AuthPolicy, Gateway | openshift-ingress |
API Group
All MaaS CRs use maas.opendatahub.io/v1alpha1 — not models.opendatahub.io/v1alpha1.
Gateway URL
MAAS_HOST=$(oc get route maas-default-gateway -n openshift-ingress -o jsonpath='{.spec.host}')
TLS=$(oc get route maas-default-gateway -n openshift-ingress -o jsonpath='{.spec.tls.termination}' 2>/dev/null)
SCHEME=$([[ -n "$TLS" ]] && echo "https" || echo "http")
MAAS_URL="${SCHEME}://${MAAS_HOST}"
The exact URL is also printed by
check_maas_availability.shanddeploy-llm.shupon success.
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- check_maas_availability.sh 17 KB runs code
- check-catalog.sh 5.3 KB runs code
- check-machines.sh 20 KB runs code
- check-vram.sh 11 KB runs code
- create-governance.sh 12 KB runs code
- create-hardware-profile.sh 8.1 KB runs code
- deploy-llm.sh 60 KB runs code
- get-default-args.sh 3.8 KB runs code
- model-configs.json 7.9 KB
- set-timeout.sh 13 KB runs code
- templates/create-api-key.sh.tmpl 2.1 KB
- templates/test-chat.sh.tmpl 1.1 KB
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.
- 11d ago First seen · 441 lines · 41 tokens per session scan A 05af5a7097eb
deploy-maas-model is a skill published in the GitHub repository MichalSteczko/rhoai-agentic-skills (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 41 tokens to every session and 4,738 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
tensorrt-llm
High-throughput LLM inference on NVIDIA GPUs.
google-cloud-solution-guided-gke-ai-migration
Guides the migration of existing AI workloads (Cloud Run, Gemini API, Gemini Enterprise Agent Platform) to self-hosted GKE inference using gcloud and kubectl. Use when the user has an existing AI inference workload (on Cloud Run, the Gemini API, Gemini Enterprise Agent Platform, or a custom VM) and wants to move it to…
agent-platform-tuning
Agent Platform Model Tuning. Use when you need to fine-tune open models or Gemini models using Agent Platform infrastructure. Don't use for model training outside Agent Platform, model deployment to endpoints (use agent-platform-deploy), or managing serving endpoints (use agent-platform-endpoint-management).
gke-inference
Deploys and optimizes AI/ML inference workloads on GKE, using GPUs, TPUs, and model servers. Use when deploying GKE inference servers, configuring GKE GPU resources for inference, or deploying LLMs on GKE. Don't use for generic batch jobs or HPC task queues (use gke-batch-hpc instead).
agent-platform-endpoint-management
Manages Agent Platform serving endpoints. Use when you need to create, list, describe, update, or delete serving endpoints for model deployment on Agent Platform. Also use when troubleshooting endpoint permission, quota, or resource busy errors. Don't use for deploying models to endpoints or for running model…
modal
Modal is a serverless cloud platform for running Python on demand, including on-demand GPUs. Use when deploying or serving AI/ML models, running GPU-accelerated workloads (training, fine-tuning, inference), serving web endpoints, scheduling batch jobs, or scaling Python code to cloud containers with the Modal SDK.