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 mouchegmouradian/claude-code-skills --skill zetic-mlangegit clone --depth 1 https://github.com/mouchegmouradian/claude-code-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/mouchegmouradian/claude-code-skills/zetic-mlange)<a href="https://agentmods.dev/skills/mouchegmouradian/claude-code-skills/zetic-mlange"><img src="https://agentmods.dev/badge/skills/mouchegmouradian/claude-code-skills/zetic-mlange/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/mouchegmouradian/claude-code-skills/zetic-mlange"><img src="https://agentmods.dev/badge/skills/mouchegmouradian/claude-code-skills/zetic-mlange.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 | $0.00155 | $0.04453 |
| Opus 5.5 | $0.00062 | $0.01781 |
| Sonnet 5.5 | $0.00031 | $0.00891 |
| Haiku 4.5 | $0.00015 | $0.00445 |
Grade A, and why
zetic-mlange 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 2mo 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 — 505 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ZETIC Melange (MLange) SDK Integration Skill
This skill helps users integrate the ZETIC Melange on-device AI SDK into mobile applications.
Naming: The product is called Melange but source code uses MLange (e.g., ZeticMLangeModel).
Key Resources
- Dashboard: https://melange.zetic.ai
- Docs: https://docs.zetic.ai
- GitHub: https://github.com/zetic-ai
- Contact: [email protected]
Prerequisites
Users need two things from the Melange Dashboard:
- Personal Key — authentication credential
- Model Name — in
account_name/project_nameformat (e.g.,"Steve/YOLOv11_comparison")
Supported upload formats: PyTorch .pt2, ONNX .onnx, TorchScript .pt
SDK Setup
Android (Gradle):
// app/build.gradle.kts
dependencies {
implementation("com.zeticai.mlange:mlange:+")
}
android {
packaging { jniLibs.useLegacyPackaging = true }
}
Add <uses-permission android:name="android.permission.INTERNET" /> to AndroidManifest.xml (SDK downloads model weights on first run).
iOS (CocoaPods):
pod 'ZeticMLange'
iOS (SPM): Add https://github.com/zetic-ai/ZeticMLangeiOS as a Swift Package dependency.
CRITICAL: Threading & Concurrency Rules
All MLange SDK calls are synchronous and blocking. They directly interface with native hardware (NPU/GPU/CPU) using fixed memory buffers. Every code example you generate MUST follow these rules:
Rule 1: Never Call on the Main/UI Thread
model.run(), model.waitForNextToken(), and even model construction can block for significant time. Always dispatch to a background thread.
Android (Kotlin):
// Always use Dispatchers.IO for blocking SDK calls (large thread pool designed for blocking work).
// Do NOT use Dispatchers.Default — it's sized to CPU cores and blocking calls would starve other coroutines.
withContext(Dispatchers.IO) {
val outputs = model.run(inputs)
}
iOS (Swift):
// Use an actor or Task to move off @MainActor
func infer() async throws -> [Tensor] {
try await Task.detached {
try self.model.run(inputs: inputs) // Runs on cooperative thread pool
}.value
}
What ships with it
2 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.
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.
- 2mo ago First seen · 505 lines · 155 tokens per session scan A 34739b381555
zetic-mlange is a skill published in the GitHub repository mouchegmouradian/claude-code-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 155 tokens to every session and 4,453 once invoked, about $0.0006 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 2026-08-31.
Other skills, from other repositories
apple-on-device-ai
Integrate on-device AI using Foundation Models framework, Core ML, and open-source LLM runtimes on Apple Silicon. Covers Foundation Models (LanguageModelSession, @Generable, @Guide, SystemLanguageModel, structured output, tool calling), Core ML (coremltools, model conversion, quantization, palettization, pruning…
vision-image-analysis-workflow
Guide Apple Vision implementation and repair for text, barcode, face, landmark, rectangle, contour, saliency, trajectory, pose, segmentation, tracking, feature-print, and image-location analysis across still images and video. Use when Apple-provided Vision requests, observations, orientation, normalized coordinates…
umap-learn
Applies UMAP-learn to nonlinear dimensionality reduction, 2D/3D embeddings, clustering preprocessing, supervised or semi-supervised UMAP, DensMAP, AlignedUMAP, and Parametric UMAP workflows.
shap
Explain and audit machine-learning predictions with SHAP. Use for selecting SHAP explainers and maskers, computing and validating feature attributions, handling multi-output explanations, and producing local or global SHAP visualizations.
optimize-for-gpu
GPU-accelerates scientific Python on NVIDIA hardware and verifies that the result is correct and faster. Use for CUDA/GPU optimization; CPU-bound NumPy, SciPy, pandas, scikit-learn, NetworkX, scikit-image, vector-search, image-processing, graph, simulation, or file-I/O workloads; CuPy, cuDF, cuML, cuGraph, cuVS…
torchdrug
Builds and troubleshoots TorchDrug 0.2.1 workflows for molecular graphs, property prediction, self-supervised pretraining, molecule generation, retrosynthesis, protein representation learning, and knowledge graph reasoning. Use when code imports torchdrug or needs its datasets, models, tasks, or Engine.