Data & AI

Fix common PyTorch bugs: percentile calculations, LayerNorm for Conv1d, buffer edge cases. Trigger when writing PyTorch code for RL or neural networks.

Convert Claude Code Subagents to OpenCode Agents format. Use when migrating from Claude Code to OpenCode or when you have existing subagent definitions to port.

Comprehensive Pharo Smalltalk development workflow guide with AI-driven Tonel editing. Provides expertise in Tonel file format syntax (class definitions with name, superclass, instVars, category, method categories, class comment placement), package structure (package.st placement, directory organization, BaselineOf dependencies), development workflow (Edit → Import → Test cycle with absolute paths, re-import timing, test execution), and Pharo best practices (CRC format documentation, method categorization conventions). Use when working with Pharo Smalltalk projects, creating or editing Tonel .st files, organizing packages and dependencies, resolving import order issues, writing class comments, implementing standard Pharo development patterns, or troubleshooting Tonel syntax.

Apply activation steering to model layers during generation

Implements TF-IDF based search engines for text datasets using vector space models and cosine similarity. Use when building search functionality, finding similar documents, ranking text by relevance, or working with text retrieval systems.

Use when creating new skills, editing existing skills, or verifying skills work before deployment - applies TDD to process documentation by testing with subagents before writing, iterating until bulletproof against rationalization

Use when enriching primary animations, adding supporting details, creating depth in motion, or making scenes feel alive without distracting from main action.

AI出力の品質を自動検証するスキル。事実確認、論理性、一貫性、幻覚（ハルシネーション）検出、バイアス分析、安全性チェックを実施し、改善提案を提供。

Help and documentation for Craft-Coder plugin.Use when: user asks about available skills, commands, or how to use this plugin.Triggers: "help", "what can you do", "craft help", "как пользоваться".

Plan and (when feasible) implement or execute user acceptance tests (UAT) /end-to-end acceptance scenarios. Converts requirements or user stories intoacceptance criteria, test cases, test data, and a sign-off checklist; suggestsautomation (Playwright/Cypress for web, golden/snapshot tests for CLIs/APIs).Use when validating user-visible behavior for a release, or mappingrequirements to acceptance coverage.

Generate properly formatted Conventional Commits messages with Claude Code footer. Use when creating git commits, writing commit messages, or the user mentions "commit", "git commit", or wants to save changes.

Build complete document knowledge bases with PDF text extraction, OCR for scanned documents, vector embeddings, and semantic search. Use this for creating searchable document libraries from folders of PDFs, technical standards, or any document collection.

Self-review checklist before marking PR ready - catch clarity, correctness, and maintainability issues with fresh eyes

Used to create a new agent. Used when a user wants to create a new agent

Modern Search Engine Optimization with GEO, AI Overviews optimization, and industry-specific strategies for E-commerce, SaaS, Healthcare, Real Estate, Restaurant, Law Firm, Finance, Travel, and Education industries

Use when working on projects that use 'beads' (bd) for issue tracking. Covers bd commands, workflow, priorities, and best practices for AI agents.

Next.js 15 + Tailwind CSS development guidelines. Modern patterns including App Router, Server/Client Components, Suspense, lazy loading, file organization, Tailwind CSS styling, performance optimization, and TypeScript best practices. Use when creating components, pages, features, styling, routing, or working with Next.js code.

Test specifications for SEO & LLM Optimizer WordPress plugin. Use when testing the SEO & LLM Optimizer plugin, verifying Copy for AI functionality, or checking markdown/RAG export features.

Stress-test book concepts against existing research before committing to architecture. Use when the user has a Book Concept Document ready for validation, wants to verify their thesis is defensible, needs to understand the competitive intellectual landscape, or wants honest assessment of their idea's strengths and weaknesses. Produces a Validation Report that informs the Go/No-Go decision. Nonfiction only.

CEO statement tone shift analyzer for /ceo-analysis page. Analyzes CEO quotes from SEC filings, earnings calls to detect tone changes (positive→cautious, cautious→aggressive) and finds historical similar patterns. Generates news articles when significant shifts detected.