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 software-ios-ai-enginegit 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/software-ios-ai-engine)<a href="https://agentmods.dev/skills/vasilyu1983/ai-agents-public/software-ios-ai-engine"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/software-ios-ai-engine/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/software-ios-ai-engine"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/software-ios-ai-engine.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 | $0.06896 |
| Opus 5 | $0.00018 | $0.03448 |
| Sonnet 5 | $0.00007 | $0.01379 |
| Haiku 4.5 | $0.00004 | $0.00690 |
Grade A, and why
software-ios-ai-engine 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 9d 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 — 245 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Local AI Engine on iOS
Use this skill when an iOS app should run useful AI behavior locally before spending cloud quota: Apple Foundation Models, deterministic local NLG, local retrieval stitching, local classifiers, extraction, summarization, tagging, rewrite helpers, and tool calls into app state. The common constraint is not "make chat smarter"; it is pick the right local engine, shape the data contract, gate capability correctly, and keep cloud as an explicit upgrade or fallback.
A major scenario is a rich, per-user structured context bundle (chart, Human Design, planning cache, knowledge chunks, activity ratings, dream themes, mood/energy, etc.) that needs to produce a real answer without cloud quota. In that case, the fix is never "make the reject card nicer." The fix is adding a local Composer tier between intent routing and cloud fallback.
This skill is for iOS product surfaces and local app-engine design. For pure retrieval/chunking/grounding strategy upstream of the local engine, route to ai-rag. For model serving/quantization tradeoffs beyond Apple platform APIs, route to ai-llm-inference. For evaluation of generated or extracted output, route to ai-evals-observer.
Quick Reference
Local Engine Patterns
| Pattern | Local engine | Best for | Fallback |
|---|---|---|---|
| Structured generation | Apple Foundation Models + @Generable |
Short prose, extraction, classification, tagging, typed transformations | Deterministic local logic or cloud opt-in |
| Deterministic NLG | Sentence bank / templates / rules | Auditable answers, safety copy, older devices, per-locale consistency | Retrieval stitch or cloud opt-in |
| Retrieval stitch | Local top-k chunks + wrappers | Grounded explanations from existing knowledge chunks | Sentence bank or cloud opt-in |
| Local classifier | Regex, NaturalLanguage, embeddings, FM enum output | Routing, intent, entity extraction, safety boundaries | Conservative default route |
| Tool-backed local model | FM tool calling into app state | Model decides when it needs app data | Pre-fetch compact data if tool overhead is too high |
| Reusable app AI foundation | LocalAIEngine facade + deterministic fallback + optional Foundation Models |
New iOS app skeletons that need AI-ready architecture before the first AI feature ships | No-op or sentence-bank engine |
| Local semantic/vector search | Natural Language embeddings, local vector table, or bundled retrieval units | User notes, settings, local knowledge, short document sets, app help, offline search | Server vector brain when corpus or sharing exceeds device scope |
What ships with it
20 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 310 B
- assets/template-foundation-models-service.md 1.4 KB
- assets/template-local-retrieval-tool.md 1.1 KB
- data/sources.json 14 KB
- evals/evals.json 2.8 KB
- learnings.consolidated.md 598 B
- learnings.md 525 B
- references/composition-with-rag-context-vector.md 28 KB
- references/foundation-models-app-skeleton.md 3.4 KB
- references/intent-router-patterns.md 12 KB
- references/local-ai-task-patterns.md 16 KB
- references/nlg-fundamentals.md 9.2 KB
- references/on-device-vector-retrieval-ios.md 3.4 KB
- references/option-a-foundation-models.md 15 KB
- references/option-b-sentence-bank.md 13 KB
- references/option-c-retrieval-stitch.md 10 KB
- references/patterns-antipatterns-traps-scenarios.md 24 KB
- references/swiftui-composer-integration.md 14 KB
- references/three-tier-architecture.md 10.0 KB
- scripts/scaffold-composers.sh 9.8 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.
- 9d ago First seen · 245 lines · 36 tokens per session scan A 0c85d0a13574
software-ios-ai-engine is a skill published in the GitHub repository vasilyu1983/AI-Agents-public (87 stars, last pushed 10d ago), licensed MIT. It adds 36 tokens to every session and 6,896 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 2026-09-03.
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.