Skill UI

An expert guide for developing production-grade AI applications, chatbots, and generative UIs using the Vercel AI SDK, Next.js, and React. Covers core concepts including streaming text (`streamText`), managing chat state (`useChat`), implementing complex tool/function calling, and ensuring robust structured data generation (JSON schemas) across multiple LLM providers.

This skill set provides a comprehensive directory of 32 production-ready engineering skills covering core development, security, AI/ML, and specialized tools. It helps users identify the appropriate expertise for tasks like architecture, frontend, backend, DevOps, and data engineering. Users should load only the specific skill needed rather than bulk-loading. Includes Python tools for analysis and scaffolding without external dependencies.

This powerful orchestrator coordinates the entire product development lifecycle by routing inquiries across 16 specialized sub-skills (e.g., RICE, OKRs, UX research, analytics). It acts as a central hub for both deterministic task routing and running continuous, machine-gated discovery loops. Use it when you need to systematically prioritize features, run A/B tests, or drive a goal from initial assumption to validated repository code.

This utility converts a single, high-level objective into a comprehensive, step-by-step build plan suitable for complex, multi-agent, or multi-session engineering projects. It automatically handles dependency mapping, detects parallel steps, and structures each step with self-contained context briefs, ensuring any new agent can execute it without prior knowledge. Ideal for large-scale refactoring or feature development spanning multiple PRs.

A comprehensive resource providing codified expertise for managing complex energy portfolios for large commercial and industrial (C&I) consumers. Covers electricity and gas procurement, tariff structure analysis (TOU, RTP), demand charge mitigation, PPA evaluation, and multi-site energy strategy development. Ideal for optimizing utility spending, understanding complex utility bills, and building resilient, sustainable energy supply chains.

A comprehensive guide to building robust, maintainable, and idiomatic Perl applications using modern features (Perl 5.36+). This resource covers best practices for object-oriented programming (Moo/Moose), advanced error handling (Try::Tiny), context sensitivity, complex regex usage, and efficient module design.

This comprehensive guide covers essential security patterns for Perl development, addressing common vulnerabilities like SQL Injection, XSS, Path Traversal, and Shell Injection. It details best practices including using Perl's Taint Mode, implementing strict allow-list validation, performing safe file operations, and executing external commands securely to ensure robust and resilient applications.

This skill is intended for Perl testing functionalities. Currently, the Japanese translation and content are under development, serving as a placeholder. Users should refer to the original English skill file for the core functionality and detailed implementation.

This skill provides sophisticated project estimation by adapting to human-only, hybrid, or agent-first development modes. It uses research-backed PERT statistics and confidence bands (P50, P75, P90) to forecast development timelines for tasks, backlogs, and capacity planning. It continuously calibrates estimates based on actual completion data, ensuring accuracy in AI-assisted and hybrid development environments.

This skill enforces rigorous governance and discipline when AI modifies systems, debugs code, or makes root-cause claims. It mandates safety gates (e.g., backup, blast radius check), demands evidence for every diagnosis, and prevents common AI shortcuts like guessing or deflecting. Use this for critical code modifications, complex debugging, and high-stakes deployments to ensure thoroughness and safety.

This command executes the entire AgentHub lifecycle in a single, chained invocation. It automates complex development tasks by sequentially handling initialization, capturing performance baselines, spawning multiple specialized agents, evaluating the results (via metrics or LLM judging), and finally presenting a winner for merging. Use this when running an end-to-end competition or complex refactoring task.

A forensic tool for performing post-mortem analysis on AI-assisted coding sessions. It goes beyond simply stating what was done, determining *why* work deviated (e.g., spec ambiguity, repo friction, agent error), identifying scope creep, and classifying rework patterns. It provides evidence-backed recommendations to improve prompts, codebase health, and future development workflows.