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-local-model-opsgit 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-local-model-ops)<a href="https://agentmods.dev/skills/vasilyu1983/ai-agents-public/ai-local-model-ops"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/ai-local-model-ops/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-local-model-ops"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/ai-local-model-ops.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 | $0.00042 | $0.02709 |
| Opus 5 | $0.00021 | $0.01354 |
| Sonnet 5 | $0.00008 | $0.00542 |
| Haiku 4.5 | $0.00004 | $0.00271 |
Grade A, and why
ai-local-model-ops 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 8d 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 — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Local Model Operations
Use this skill to choose and operate local or self-hosted LLM workflows when privacy, offline access, or low-friction experimentation matter more than large-cluster serving.
This skill covers:
- local runtime choice for laptops, workstations, and small self-hosted setups
- team-facing local or private chat surfaces
- single-binary or minimal-dependency model packaging
- lightweight adaptation paths before full training or cluster-scale serving
- evaluation and escalation rules before a local stack becomes a product dependency
ASCII Flow
local/private model need
|
v
constraint
privacy | offline | cost | hardware | demo portability | team chat
|
v
runtime selection
Ollama | LM Studio | MLX | Microsoft Foundry Local | Open WebUI | llamafile | lightweight adapter workflow
|
v
local operating contract
pinned model + quantization + eval set + storage/privacy boundary
+ optimization levers: KV-cache quant | speculative decoding | NPU tier
|
v
use or escalate
local workflow OR hand off to inference/MLOps for production serving
Quick Reference
| Need | Default path | Notes |
|---|---|---|
| Run a local model quickly | Ollama | Lowest-friction day-0 local runtime for experiments and private workflows |
| Share a self-hosted chat UI | Open WebUI | Best fit when a team needs a ChatGPT-like local or private interface |
| Ship a no-install demo or portable binary | llamafile | Useful for single-file distribution and low-ops delivery |
| Apple Silicon on-device inference at framework level | MLX (mlx-lm) | Primary path for Metal-native inference and LoRA fine-tune on Mac; verify at https://github.com/ml-explore/mlx-lm |
| GUI model browser and switcher (non-technical users) | LM Studio | Supports GGUF and MLX; good for rapid model comparisons |
| Windows / enterprise SDK-first local inference | Microsoft Foundry Local | Curated Microsoft catalog; SDK + REST; verify at https://learn.microsoft.com/en-us/ai/foundry-local |
| Fine-tune or adapt cheaply | Unsloth + ../ai-llm/SKILL.md |
Good for lightweight adaptation, not a substitute for full training ops |
| Optimize throughput or production serving | ../ai-llm-inference/SKILL.md |
Use this skill for local ops; use ai-llm-inference for deeper serving engineering |
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.
- agents/openai.yaml 368 B
- assets/templates/ollama-setup-recipe.md 5.2 KB
- assets/templates/openwebui-deployment-recipe.md 5.3 KB
- data/sources.json 5.8 KB
- learnings.consolidated.md 594 B
- learnings.md 460 B
- references/adaptation-and-packaging.md 1.4 KB
- references/desktop-runtime-landscape.md 5.9 KB
- references/model-sizing-matrix.md 19 KB
- references/quantization-format-table.md 11 KB
- references/runtime-selection.md 1.4 KB
- references/small-model-tier-table.md 7.6 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.
- 8d ago Changed · +1 lines 494b7aa953d7
- 12d ago First seen · 173 lines · 42 tokens per session scan A d23a873b7c1c
ai-local-model-ops is a skill published in the GitHub repository vasilyu1983/AI-Agents-public (87 stars, last pushed 9d ago), licensed MIT. It adds 42 tokens to every session and 2,709 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.