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 vasilyu1983/AI-Agents-public --skill ai-mlopsgit clone --depth 1 https://github.com/vasilyu1983/AI-Agents-publicWrote 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/vasilyu1983/ai-agents-public/ai-mlops)<a href="https://agentmods.dev/skills/vasilyu1983/ai-agents-public/ai-mlops"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/ai-mlops/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/vasilyu1983/ai-agents-public/ai-mlops"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/ai-mlops.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 | $0.00036 | $0.05309 |
| Opus 5 | $0.00018 | $0.02655 |
| Sonnet 5 | $0.00007 | $0.01062 |
| Haiku 4.5 | $0.00004 | $0.00531 |
Grade A, and why
ai-mlops 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 7d 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 — 286 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MLOps & LLMOps - Production Operations Hub
July 2026 posture: version every changeable artifact, gate every release with a regression-eval suite in CI, instrument the whole path with OpenTelemetry (pin GenAI convention version — the spec now lives in its own repo and is still evolving), treat tool/RAG context as untrusted input, and ship rollback plus incident playbooks before launch.
This skill is the execution hub for operating AI systems in production:
- Classical ML ops: ingestion, registries, feature stores, drift, retraining, promotion
- LLMOps: serving, prompt/config lifecycle, online evals, cost controls, safety gates
- Agent runtime ops: tracing, tool governance, approval paths, MCP-aware telemetry, rollback
- Governance: privacy, supply chain, auditability, AI Act readiness, safety incident handling
Use this skill for production architecture, release gates, monitoring, incidents, and governance. Use adjacent skills for modelling, retrieval depth, agent design, or inference internals.
When To Use This Skill
Activate this skill when the user asks for:
- Deploying an ML, LLM, RAG, or agent-backed system to production
- Designing serving, batch, hybrid, or multi-region runtime architecture
- Adding observability, drift detection, alerting, retraining, or release gates
- Writing incident runbooks, rollback plans, or go/no-go checklists
- Hardening an AI system against prompt injection, RAG poisoning, tool abuse, or data leakage
- Building governance artifacts for privacy, auditability, or regulated rollout
- Choosing how to operate prompts, model artifacts, feature definitions, or agent graphs safely
- Diagnosing why changes to an ML system keep rippling: entanglement, correction cascades, undeclared consumers, pipeline jungles, or config sprawl
- Operating fairness, privacy-budget, human-oversight, appeal, watermark/provenance, copyright/memorization, or environmental controls
- Deploying multimodal image, document, audio, video, vision-language, or diffusion systems with bounded media ingestion, safety, latency, and cost
What ships with it
53 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.
- agents/openai.yaml 322 B
- assets/deployment/deployment-readiness-checklist.md 7.0 KB
- assets/deployment/template-api-service.md 1.5 KB
- assets/deployment/template-batch-pipeline.md 1.4 KB
- assets/deployment/template-deployment-mlops.md 1.8 KB
- assets/governance/template-policy-checklist.md 835 B
- assets/governance/template-risk-assessment.md 701 B
- assets/governance/template-security-audit.md 1.0 KB
- assets/incident/template-incident-runbook-safety.md 1.1 KB
- assets/incident/template-jailbreak-investigation.md 1.0 KB
- assets/monitoring/template-drift-retraining.md 1.1 KB
- assets/monitoring/template-monitoring-plan.md 1.2 KB
- assets/ops/template-incident-runbook.md 1.6 KB
- assets/privacy/template-data-anonymization.md 784 B
- assets/privacy/template-pii-handling.md 877 B
- assets/safety/template-guardrail-config.md 1.4 KB
- assets/safety/template-output-filter.md 917 B
- assets/safety/template-safety-prompt.md 983 B
- data/sources.json 19 KB
- learnings.consolidated.md 584 B
- learnings.md 485 B
- references/agentic-security.md 19 KB
- references/agentops-patterns.md 13 KB
- references/api-design-patterns.md 3.3 KB
- references/automated-retraining-patterns.md 22 KB
- references/cost-management-finops.md 19 KB
- references/data-ingestion-patterns.md 14 KB
- references/deployment-lifecycle.md 18 KB
- references/deployment-patterns.md 3.9 KB
- references/drift-detection-guide.md 8.0 KB
- references/edge-mlops-patterns.md 14 KB
- references/experiment-tracking-patterns.md 16 KB
- references/extraction-defense.md 17 KB
- references/feature-store-patterns.md 29 KB
- references/governance-checklists.md 3.8 KB
- references/incident-response-playbooks.md 2.8 KB
- references/jailbreak-defense.md 1.4 KB
- references/llm-rag-production-patterns.md 24 KB
- references/ml-technical-debt-taxonomy.md 27 KB
- references/model-registry-patterns.md 20 KB
- references/monitoring-best-practices.md 3.6 KB
- references/multi-region-patterns.md 20 KB
- references/online-evaluation-patterns.md 29 KB
- references/output-filtering.md 13 KB
- references/privacy-protection.md 1.5 KB
- references/prompt-injection-mitigation.md 1.7 KB
- references/rag-security.md 13 KB
- references/responsible-multimodal-operations.md 11 KB
- references/safety-evaluation.md 16 KB
- references/supply-chain-security.md 11 KB
- references/threat-models.md 3.7 KB
- scripts/deployment_smoke_test.sh 6.7 KB runs code
- scripts/drift_check.py 6.1 KB runs code
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.
- 7d ago Changed · +5 lines · -4 tokens per session 9289ebfac6d2
- 11d ago First seen · 281 lines · 40 tokens per session scan A c98bae196440
ai-mlops is a skill published in the GitHub repository vasilyu1983/AI-Agents-public (87 stars, last pushed 9d ago), licensed MIT. It adds 36 tokens to every session and 5,309 once invoked, about $0.0002 per session on Opus 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
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-memory-thermal-ops
Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
data-quality-frameworks
Implement data quality validation with Great Expectations, dbt tests, and data contracts. Use when building data quality pipelines, implementing validation rules, or establishing data contracts.
airflow-dag-patterns
Build production Apache Airflow DAGs with best practices for operators, sensors, testing, and deployment. Use when creating data pipelines, orchestrating workflows, or scheduling batch jobs.
dbt-transformation-patterns
Master dbt (data build tool) for analytics engineering with model organization, testing, documentation, and incremental strategies. Use when building data transformations, creating data models, or implementing analytics engineering best practices.