Skill UI

Provide step-by-step explanations of specific files, functions, or modules, leveraging the understand-explain workflow to map graph context, layer placement, internal structure, and data flow for developers needing deep clarity.

Turns messy workflows, docs, transcripts, or existing scripts into production-ready cross-platform agent skills by surfacing hidden requirements, generating specs, and validating exports for Claude Code, Copilot CLI, Cursor, and similar runtimes.

A comprehensive guide and set of scripts to bootstrap, build, package, sign, and notarize SwiftPM-based macOS applications without requiring Xcode projects. It provides a complete end-to-end workflow for managing app releases, including generating AppCasts and handling complex code signing and notarization steps.

This skill provides a comprehensive audit workflow for diagnosing performance issues in SwiftUI applications. It begins with a code-first review, identifying common performance smells like invalidation storms or layout thrash. If code inspection is insufficient, it guides the user through collecting runtime profiling evidence (e.g., Instruments data) to pinpoint root causes such as janky scrolling, high CPU usage, or memory leaks.

This comprehensive workflow guides the design, optimization, and deployment of scalable, cost-effective Google Cloud Platform (GCP) architectures for both startups and large enterprises. It covers the full lifecycle, from gathering detailed requirements (compliance, SLA) and recommending optimal patterns (Serverless, GKE, Data Pipelines) to estimating costs, generating Infrastructure as Code (IaC) using Terraform, setting up CI/CD pipelines, and completing a thorough security review.

This guide enforces the 'spec-first' principle in software development, treating the specification as a non-negotiable contract. It outlines a rigorous, multi-stage workflow detailing mandatory sections (Functional/Non-Functional Requirements, Acceptance Criteria, API Contracts, Data Models) and bounded autonomy rules. Users learn how to define the system 'must do' before writing a single line of code, thereby minimizing rework and eliminating scope creep.

Workspace Surface Audit scans repo files, MCP servers, plugins, connectors, and environment configs to report current capabilities and recommend ECC-native skills, hooks, agents, or workflows for setups like Claude Code.

Accelerates CPU-bound Python workflows by translating NumPy, pandas, ML, graph, image, simulation, and IO code into NVIDIA GPU kernels using CuPy, Numba CUDA, Warp, cuDF/cuML/cuGraph, KvikIO, cuSpatial, and RAFT for dramatic speedups.

Extracts business domain knowledge from a codebase, optionally reusing an existing knowledge graph, and generates an interactive horizontal flow graph for dashboard visualization, with lightweight scanning and validation steps.

Guides creation of persona-targeted VS Code CodeTour .tour files, covering repo exploration, intent inference, step composition, and validation so each narrative walkthrough is anchored to real files and tailored depth.

Keeps structured JSON records for every code change, enforces a state-machine workflow, and hands off AI sessions with summaries, blockers, and files so bots can resume seamlessly after interruptions.

Guides applying Dart 3 branches, patterns, pattern types, and records when writing Flutter/Dart code, covering if-case, switch expressions, guards, and record usage to encourage correct, exhaustive handling in control-flow and destructuring scenarios.