ml-optimization
Optimizes TensorFlow.js WebGPU backend, WebLLM offline inference, and implements memory safety for heavy ML models
Découvrez les skills Bases de données, comparez leurs dépôts et consultez leurs instructions.
Optimizes TensorFlow.js WebGPU backend, WebLLM offline inference, and implements memory safety for heavy ML models
Normalizes image histograms to preserve melanin features using Canvas API before analysis
Validates image file signatures using Magic Bytes and calculates SHA-256 cryptographic hash
Commits transaction hash to immutable AgentDB ledger for complete audit trail of clinical analysis
Extracts vector embeddings for fairness analysis using MobileNetV3 with FairDisCo disentanglement
Isolates skin regions from background clutter using calibrated thresholds for downstream analysis
Retrieves 10 similar historical cases from AgentDB vector store for RAG-based diagnosis support
Central Goal-Oriented Action Planning orchestrator that plans and executes clinical analysis pipelines
Applies decision thresholds for high-confidence inputs or enforces conservative safety margins for low-confidence cases
Manage Git worktrees for working on multiple branches simultaneously without cloning the repository
Classifies skin conditions including Melanoma and Basal Cell Carcinoma using TF.js MobileNetV3
Synthesizes lesion data and historical cases into a comprehensive risk profile using WebLLM SmolLM2
Grounds clinical diagnosis via live medical literature search using Google Search Tool
Classifies patient skin tone using Fitzpatrick/Monk scale with Gemini 3 Flash and outputs confidence score
Create GitHub releases with version tags, release notes, and changelog generation from merged PRs
Generates actionable clinical advice (max 25 words) calibrated for patient skin tone using WebLLM or Gemini
Updates local AgentDB vector store with new case patterns while monitoring for bias introduction
Implements comprehensive test suite using Vitest for unit tests and Playwright for E2E tests with safety interception scenarios
Create, amend, and manage Git commits with gh CLI including status checks and commit message validation
Debug JavaScript/TypeScript applications using browser DevTools protocol, VS Code debugger, or Node.js inspector
Manages CI/CD pipelines, linting with ESLint/SonarJS, and release tagging for production deployments
Validates True Positive Rate and False Positive Rate gaps across demographics using AgentDB metrics
Enforces CSP, input sanitization, encryption standards, and GDPR compliance for clinical data
Validate translation keys exist in both Polish and English locale files before committing. Use when adding new user-facing text or translation keys.
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.