Skill UI

An autonomous factory for generating production-ready, cross-platform AI agent skills. It accepts raw, unstructured human input—such as transcripts, PDFs, screenshots, vague ideas, or code snippets—and reverse-engineers the complete workflow logic. It bridges the gap between human intent and functional agent specifications, allowing users to automate complex processes without writing code.

This agent facilitates comprehensive debugging and testing of iOS applications running on a simulated environment. It leverages XcodeBuildMCP to manage the entire lifecycle, including building, running, interacting with the UI (tap, type, gesture), taking screenshots, and capturing detailed runtime logs. Use it whenever you need to simulate real-world app testing or diagnose issues on a booted simulator.

A comprehensive toolset to scaffold, build, sign, and package SwiftPM-based macOS applications entirely without requiring an Xcode project. It guides users through the end-to-end process, including setting up the project, running the app, performing code signing, notarization, and generating update appcasts for robust, reproducible releases.

This skill provides a comprehensive, structured workflow for diagnosing performance bottlenecks in SwiftUI applications. It guides users through a code-first review to detect common issues like invalidation storms, layout thrash, or heavy main-thread work. When code review is insufficient, it provides detailed guidance on collecting runtime evidence using Instruments, enabling precise root cause analysis and targeted remediation strategies.

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.

A comprehensive, library-agnostic checklist covering best practices for Flutter and Dart applications. It guides developers on architectural principles, state management patterns (BLoC, Riverpod, etc.), Dart idioms, performance optimization, and clean code principles, making it invaluable for code reviews.

This tool monitors Modbus TCP communication within SCADA and ICS environments to detect security anomalies. It performs deep packet inspection, baselines normal communication patterns (e.g., typical function code distribution, register access), and applies statistical analysis to identify unauthorized register writes, reconnaissance attempts, or denial-of-service attacks targeting industrial control devices.

repo-scan is a powerful tool designed to audit large, complex, cross-stack codebases (C++, Java, iOS, Web). It classifies every file into project code, third-party dependencies, or build artifacts. Key features include detecting embedded libraries (e.g., FFmpeg, OpenSSL) and providing a practical 4-stage assessment (Core Asset / Extract / Refactor / Deprecate) via an interactive HTML report. Essential for managing monorepos and planning major refactoring efforts.

The Safety Guard skill is designed to prevent destructive actions when working in production systems or when autonomous agents are executing tasks. It offers three protection modes (Careful, Freeze, and Guard) to monitor dangerous commands, lock specific directories, and ensure code changes are confined to intended areas, enhancing system stability and security.

This skill detects malicious Command and Control (C2) communications tunneled through the DNS protocol. It analyzes DNS query logs for indicators such as DNS tunneling tools (Iodine, dnscat2), Domain Generation Algorithms (DGA), and encoded payloads within TXT/CNAME records. Techniques include Shannon entropy analysis of subdomains, statistical anomaly detection, and machine learning classification. Ideal for network security analysts conducting deep threat hunting and forensic investigations.

This harness transforms standard code environments into fully persistent, autonomous agent systems. It leverages native features like scheduled tasks (Crons), remote dispatching, and structured memory storage (MCP) to enable continuous, self-directing automation. Ideal for building AI assistants that require long-term context, recurring workflows, and complex, multi-step operations, eliminating the need for external agent frameworks like AutoGPT.

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.