Skill UI

This tool executes the foundational SPARC Specification Phase (Phase 1) for new features or projects. It guides the user through systematically gathering comprehensive requirements, including functional/non-functional needs, defining concrete acceptance criteria (Given/When/Then), identifying technical constraints, and mapping edge cases. It creates a robust, formalized specification document that serves as the blueprint for all subsequent design and development phases.

A comprehensive guide to building high-performance, scalable UIs using the latest SwiftUI best practices. Covers advanced state management with @Observable, type-safe navigation using NavigationStack, view composition techniques, and crucial performance optimization tips for iOS and macOS development.

This skill executes comprehensive historical backtesting using the high-performance neural-trader engine. It supports features like walk-forward validation and captures key financial metrics (Sharpe, Drawdown). Critically, it signs the resulting artifact using Ed25519 cryptography, providing non-repudiable tamper evidence required for promoting research models to live trading systems.

A comprehensive tool for training and evaluating advanced neural network models (including LSTM, Transformer, and N-BEATS) on time-series market data. It facilitates the prediction of financial trends, generating results with confidence intervals, and comparing performance across multiple architectures. Ideal for quantitative trading research and financial forecasting.

Tree Ring Memory is a local-first, framework-agnostic layer for managing durable knowledge in AI agents. It provides structured methods for recalling past decisions, recording evidence, auditing project state, and deliberately forgetting information, ensuring the core project memory remains clean and actionable without relying on raw conversation transcripts.

This comprehensive tool automates the process of translating non-English official documents (images) for visa applications. It handles image preprocessing (conversion and rotation), performs robust OCR using multiple frameworks, translates the extracted text into professional English while maintaining document structure, and generates a certified bilingual PDF containing both the original image and the formatted translation.

This tool provides a comprehensive overview of your project's memory usage across all associated memory systems. It reports metrics such as line counts, topic file counts, capacity limits, stale entries, and actionable recommendations to help maintain memory integrity and optimize performance.

Executes a multi-agent, structured code review workflow for GitHub Pull Requests (PRs). It analyzes changes from multiple perspectives (e.g., bug scans, compliance with project guidance, git history, etc.), provides confidence scoring, and generates actionable, highly formatted feedback directly on the PR. Use this when a thorough, professional code review is required.

This tool generates a detailed, local 'receipt' or impact report of all Claude Code activity recorded in local session transcripts. It cross-references this data with your local Git history to provide a quantifiable measure of effort and project contribution. Ideal for performance reviews, budget justifications, or self-auditing usage.

This skill provides a robust framework for self-certifying task delivery. It guides developers through discovering a formal "verify plan" and capturing concrete evidence—such as screenshots, DOM snapshots, or CLI output—across defined surfaces (Web, CLI, Electron). The process ensures that task delivery is judged on demonstrable, reproducible proof, eliminating reliance on mere textual claims.

This skill connects to the public, read-only AI HOT API to retrieve the latest Chinese AI news, curated selections, hot topics, and daily reports. It is essential for users needing up-to-date information on large language models, industry breakthroughs, or major AI events, ensuring results are based on real-time data rather than pre-trained knowledge.