メニュー
Create or update a skill and mirror it to both .github/skills/ and .claude/skills/
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
""Optimize and deploy ML inference on edge devices. Use when: running LiteRT on Snapdragon XR1, configuring GPU/NNAPI delegates, optimizing memory for 6GB RAM constraint, reducing inference latency below 50ms, deploying Gemma 4 on phone, running MLX on Apple Silicon, profiling thermal throttling, battery-aware inference scheduling.""
""Quantize and optimize ML models for edge deployment across all four tiers. Use when: converting models to LiteRT, ONNX export, INT8/FP16 quantization, GGUF conversion, MLX format, BitNet 1-bit quantization, model size reduction, inference optimization, Adreno GPU delegate, NNAPI configuration, profiling model performance on target hardware.""
""Fine-tune and train ML models using Unsloth Dynamic QLoRA for the Duchess platform. Use when: fine-tuning Gemma 4, training Qwen2.5-VL, creating domain adapters (Safety, Spanish, Engineering, Electrical), configuring QLoRA hyperparameters, running training on RTX 5090, dataset preparation for construction safety, model evaluation against benchmarks.""
""Build Android apps for Vuzix M400 glasses and companion smartphones. Use when: developing Kotlin Android code, AOSP without Google Play Services, Vuzix SDK integration, companion phone app, BLE communication, camera pipeline, sideloading APKs, foreground services, Jetpack Compose UI, Android Gradle builds, min SDK 33 for glasses, min SDK 30 for phone.""
""Design AR interfaces for Vuzix M400 glasses and companion phone app. Use when: designing heads-up display layouts, 640x360 OLED optimization, voice UI commands, safety alert presentation, glove-compatible interactions, bilingual UI layout, notification design, wearable ergonomics, OLED power optimization, accessibility for construction workers.""
""English/Spanish localization for construction safety communication. Use when: translating safety alerts, creating bilingual UI strings, construction-register Spanish terminology, code-switching training data, regional Spanish variants, OSHA bilingual compliance, LoRA adapter vocabulary for Spanish construction jargon, toolbox talk translations.""
| description | Create or update a skill and mirror it to both .github/skills/ and .claude/skills/ |
Create a new Duchess skill or update an existing one, automatically mirrored to both AI tool directories.
/skill <skill-name> [description]
Example:
/skill edge-inference "Optimize and deploy ML inference on edge devices with LiteRT and ONNX"
/skill ppe-detection-pipeline "End-to-end PPE pipeline from glasses camera to worker alert"
.github/skills/<skill-name>/SKILL.md with the Duchess SKILL formatscripts/mirror-skills.sh github to mirror it to .claude/skills/<skill-name>.md.claude/settings.json to register the new skill.github/skills/<name>/SKILL.md (canonical)---
name: <name>
description: "<one-line description>"
---
# <Name>
## When to Use
- ...
## Procedure
1. ...
## Key Patterns
...
## Pitfalls
...
.claude/skills/<name>.md (auto-generated mirror)Generated automatically by mirror-skills.sh from the .github canonical.
Do not edit the .claude copy directly — edits will be overwritten on next sync.
| Source | Destination | When |
|---|---|---|
.github/skills/*/SKILL.md | .claude/skills/*.md | Skill created/updated in .github |
.claude/skills/*.md | .github/skills/*/SKILL.md | Skill created directly in .claude |
| Either | Both | scripts/mirror-skills.sh auto |
The pre-commit hook runs mirror-skills.sh check — commits are blocked if skills are out of sync.
The post-merge hook runs mirror-skills.sh auto — pulls automatically keep skills in sync.
# After creating a .github/skills/<name>/SKILL.md manually:
./scripts/mirror-skills.sh github
# After creating a .claude/skills/<name>.md manually:
./scripts/mirror-skills.sh claude
# Check for drift without modifying anything:
./scripts/mirror-skills.sh check