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 zorost/AI-Engineering-Lab --skill cloud-deploy-gategit clone --depth 1 https://github.com/zorost/AI-Engineering-LabWrote 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/zorost/ai-engineering-lab/cloud-deploy-gate)<a href="https://agentmods.dev/skills/zorost/ai-engineering-lab/cloud-deploy-gate"><img src="https://agentmods.dev/badge/skills/zorost/ai-engineering-lab/cloud-deploy-gate/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/zorost/ai-engineering-lab/cloud-deploy-gate"><img src="https://agentmods.dev/badge/skills/zorost/ai-engineering-lab/cloud-deploy-gate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00049 | $0.01052 |
| Opus 5.5 | $0.00020 | $0.00421 |
| Sonnet 5.5 | $0.00010 | $0.00210 |
| Haiku 4.5 | $0.00005 | $0.00105 |
Grade A, and why
cloud-deploy-gate 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 1mo 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cloud Deploy Gate
1 · Purpose
Make "deploy to the cloud platform" a checklist with an owner, so the same agent that passed locally cannot become an unbudgeted, unguarded, unowned cloud workload.
2 · When to use
- Before deploying any model endpoint, RAG app, or agent to Azure AI Foundry, Google Vertex AI, AWS Bedrock, or Databricks Model Serving.
- When a cloud deployment already exists and nobody can answer the gate questions.
3 · Inputs
- The working local artifact with its eval scores and traces.
- The target platform account with billing access visible.
- The spec's refused tradeoffs (they now become platform settings).
4 · Procedure
- Pack the evals. The golden set and scorer travel with the deployment. First action in the cloud: re-run the eval there. The cloud score must match the local score within noise before anything else proceeds.
- Set the budget alarm before the first request. Every platform has billing alerts; set one at the week's number (e.g. $25) and one at the panic number (e.g. $100). An unbudgeted experiment is how tutorials become invoices.
- Turn on the platform guardrails mapped to the spec. Foundry content filters, Bedrock Guardrails, Vertex safety settings, map each refused tradeoff to its setting, in writing.
- Pin the model version. Deploy an explicit model version/ARN, not "latest". Record it in the spec.
- Scope the credentials. A dedicated service account / managed identity with least privilege: only the data sources this deployment reads. No personal credentials in any deployment.
- Enable logging and tracing at the platform level (invocation logs, prompt/ response logging where policy allows) and verify one request shows up end-to-end.
- Smoke-test the four failure modes: empty retrieval, over-long input, rate-limit, and guardrail block. Each must produce the designed response, not a stack trace.
- Name the owner in the deployment description/tag: the human who gets the budget and guardrail alerts. No orphan deployments.
- Write the teardown commands before leaving the page, delete endpoint, release resources, revoke credentials. A training deployment lives exactly as long as its exercise unless its owner says otherwise.
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.
- 1mo ago First seen · 96 lines · 49 tokens per session scan A 245e96e0b556
cloud-deploy-gate is a skill published in the GitHub repository zorost/AI-Engineering-Lab (317 stars, last pushed 2mo ago), licensed MIT. It adds 49 tokens to every session and 1,052 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
hugging-face-model-trainer
This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV…
fine-tuning-os
Use this skill when conducting a professional LLM fine-tuning engagement in Zero-Data mode via the MCP server fine-tuning-os. Triggers on: prestation de fine-tuning LLM, livraison de modèle affiné, pipeline Zero-Data, orchestration des 64 outils fine-tuning-os, cycle de vie entraînement (préparation / données…
torchtitan
Pretrain LLMs at scale with PyTorch 4D parallelism.
modal
Serverless GPU cloud for ML jobs and model APIs.
tensorrt-llm
High-throughput LLM inference on NVIDIA GPUs.
ai-engineering-toolkit
Overview of six AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.