Research

Iteratively refine LA-Bench experimental procedures through validation and regeneration cycles. This skill should be used when improving generated procedures by ID, validating procedure quality against LA-Bench criteria, and managing the refinement loop between procedure-checker and procedure-generator skills. Triggered by requests to refine, improve, or validate LA-Bench procedures in output JSONL files.

Research events and conditions data for Catskill Crew newsletter. Use when gathering Instagram events, fetching sun/moon/weather data, or preparing data for newsletter writing.

Research best practices, guides, and libraries. Use when needing documentation, recommendations, implementation guidance, or library comparisons.

Adversarial quality gate agent for code review - finds flaws before users do

Generate detailed research budgets and project timelines for grants, proposals,evaluations, and consulting. Provides NIH/NSF-compliant budgets, multi-tier scenarios,and Gantt-style timelines with UNC-specific rates.USE WHEN user says "create budget", "generate timeline", "project budget","research budget", "grant budget", "proposal budget", "cost estimate","work plan", "Gantt chart", or requests budget/timeline planning.

Generate images using Google Gemini NanoBanana via browser automation. Use this skill for general-purpose AI image generation from text prompts. Includes persistent authentication, automatic environment setup, and reference image support for style matching.

This skill should be used when investigating .NET project dependencies, understanding why packages are included, listing references, or auditing for outdated/vulnerable packages.

Use for Phase 2 of Course OS - conducting comprehensive topic research using a 4-pass strategy (landscape, deep dive, gap filling, synthesis) to build complete subject matter understanding. Triggers on "/course-research", "research topic", "deep dive research", or after completing Phase 1.

Use this skill to discover all available SynthesisFlow skills and their capabilities. Provides a bootstrap context for AI agents by listing all skills, their descriptions, and script paths from the .claude/skills/ directory.

Create psychologically compelling, accessibility-grounded frontend interfaces using avant-garde design concepts not yet mainstream in 2025. Integrates shadcn registries, experimental layout systems beyond traditional grids, and design patterns that elicit visceral "WOW" responses. Use when building React/Next.js components, landing pages, dashboards, or any interface requiring distinctive visual identity with biomorphic aesthetics, kinetic typography, soft tactile UI, or computational organic forms. Triggers on requests for unique design systems, custom shadcn themes, registry integration, or when users want interfaces that feel genuinely designed rather than templated.

Use when justifying technical debt to executives, calculating the cost of quality issues, translating tech metrics to business language, or planning quality budgets - uses research-backed formulas (2-3x defects, productivity multipliers) to convert code problems into dollars and ROI

Takes user stories and enriches them with implementation details, patterns, and file references. Spawns architect to add API endpoints, database models, component names, and cross-cutting concerns. Use when converting high-level stories to architecture-annotated specifications ready for execution.

Strategic design agent focused on codebase research and implementation blueprinting.

Scrape external documentation (API references, library docs, protocol specifications) and generate Claude Agent Skills in SKILL.md format. Use when creating skills from documentation, integrating third-party knowledge, or building domain-specific skills. Parses HTML/Markdown, extracts structure, generates proper frontmatter and instructions. Triggers on "scrape documentation", "docs to skill", "generate skill from docs", "API documentation to skill".

Conducts rigorous academic research and document analysis using Gemini's advanced control features (Logprobs, Citations, Long Context).

PostgreSQL conventions overview for Supabase projects. Use when: (1) Need overview of database design principles, (2) Understanding tight-fit design philosophy, (3) Quick reference for standard table structure

Storybook CSF3 story authoring methodology for UI component catalogs.Prioritizes visual showcase patterns over exhaustive argTypes enumeration.Reference for creating scannable, maintainable component documentation.

Conducts adaptive multi-round interviews to initialize learning resource projects. Collects topic, domain, motivation, background, preferences, and volume through dynamic conversation using ARCS/SDT theories. Use when starting new projects via /init-project or reinitializing project configuration.

Cross-database search using multiple academic sources. Triggers: 跨資料庫, multi-source, Semantic Scholar, OpenAlex, CORE, Europe PMC, 綜合搜尋

Capture important decisions made during Claude Code sessions and store them in searchable, structured format for future reference. Automatically triggers on decision keywords or manual request.