skills.rest
maziyarpanahi
maziyarpanahiCOMMUNITY
Explore all skills by this author →
Website𝕏@maziyarpanahiGitHub

pick-a-pii-model

Select an on-device PII detection model by language, runtime format, and device size.

5.0k|615|Updated Oct 4, 2025
Search Tags:#model selection#model registry#on-device ai#mobile deployment#pii detection#offline inference#recall validation
One-click install
npx skills add https://github.com/maziyarpanahi/openmed --skill pick-a-pii-model
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pick-a-pii-model
Source: https://github.com/maziyarpanahi/openmed/tree/main/skills/pick-a-pii-model
Command: npx skills add https://github.com/maziyarpanahi/openmed --skill pick-a-pii-model

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps select an appropriate on-device personally identifiable information model without relying on live model discovery, while preventing deployment decisions based solely on size or runtime compatibility.

Core Features & Use Cases

  • Offline Model Shortlisting: Filters the committed PII model registry by language, runtime format, and device size budget.
  • Safety-Oriented Selection: Preserves language defaults, rejects unsupported languages, and requires recall validation before deployment.
  • Deployment Guidance: Supports CPU, GPU, Apple Silicon, mobile export, Android, Core ML, ONNX, and browser workflows through compatible source formats.
  • Use Case: Choose an English PII detector for Apple Silicon under a defined memory budget, then benchmark direct-identifier recall and critical leakage before release.

Quick Start

Use the pick-a-pii-model skill to shortlist a compatible local PII model for the target language, runtime, and device budget, then validate its recall before deployment.

Frequently Asked Questions about pick-a-pii-model

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose an on-device PII detection model for offline mobile deployment?▼

To choose an on-device PII detection model, filter a committed model registry by target language, runtime format like ONNX or Core ML, and device memory size budget. This ensures compatible offline inference without relying on live model discovery.

What is the safest way to select an offline PII model for Apple Silicon?▼

Safe PII model selection requires filtering by compatible source formats for Apple Silicon and preserving language defaults. You must also conduct post-conversion recall validation for direct identifiers and critical leakage before releasing the model.

Can I deploy a PII detection model for Android using ONNX runtime?▼

Yes, you can deploy a PII detection model for Android using ONNX by filtering the model registry for ONNX-compatible source formats. The selection process requires measuring local resource constraints and validating identifier recall before final deployment.

Why should I not select an on-device PII model based solely on model size?▼

Selecting an on-device PII model based solely on model size risks unsafe deployments and critical data leakage. Safe selection requires filtering by language defaults, runtime compatibility, and post-conversion recall validation to ensure accurate offline inference.

What are the limitations of using a committed registry for offline PII model selection?▼

A committed registry prevents live model discovery but requires rejecting unsupported languages entirely. You must manually apply compatible format filtering for CPU, GPU, or browser workflows and validate recall against direct identifiers before any release.

Related Skills

Explore More Software Engineering Skills→
AG064AG064
2

sherpa-onnx-tts

Convert input text to speech offline using sherpa-onnx runtime.

Community
Intermediate
maziyarpanahimaziyarpanahi
5.0k

evaluating-with-leakage-gates

Evaluate de-identification and clinical NER models for residual PHI leakage before release.

Community
Advanced
ovachieverovachiever
52

llama-cpp

Run LLM inference on CPU and non-NVIDIA hardware with GGUF quantization.

Community
Advanced
maziyarpanahimaziyarpanahi
5.0k

choosing-openmed-models

Select OpenMed clinical or biomedical models via structured ModelQuery filters.

Community
Intermediate