Research

Read-only code exploration and documentation search agent

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

Use this skill for early-stage project planning through Product Requirements Documents (PRDs). Guides users from initial project ideas through product briefs, market research, PRD creation, validation, and epic decomposition. Triggers include "create PRD", "product brief", "validate requirements", or beginning project inception activities.

Research accelerator portfolio companies using Firecrawl and Tavily MCPs. Generates structured CSV and markdown reports with systematic impact scoring. Optimized for token efficiency.

Modern CSS styling methodology prioritizing CSS Grid over Flexbox.Covers Grid layout patterns, Subgrid, Container Queries, Design Tokens, and maintainable CSS architecture.Reference for implementing scalable, responsive layouts with predictable behavior.

Track progress over time across therapy sessions and journal entries. Supports trend analysis, session comparisons, period summaries, and goal tracking. Reads from analysis files and generates comprehensive progress reports.

Competitive analysis framework for CollectiveGood positioning. Researches companies in AI validation,human-in-the-loop systems, clinical AI evaluation spaces. Produces comprehensive competitive assessmentscovering business model, technology, market positioning, and differentiation opportunities.USE WHEN user says "competitive analysis", "assess competitor", "compare to [company]","market positioning", or requests research on companies in CollectiveGood's space.

Use before committing staged changes when you need to verify all related documentation is current - systematically checks README, CLAUDE.md, CHANGELOG, API docs, package metadata, and cross-references rather than spot-checking only obvious files

Multi-agent development workflow system for structured task execution. Use when working with tasks in .tasks/ directory, when user invokes /task, /explore, /research, /spec, /build, /review, or /refactor commands, or when Claude needs to understand the task orchestration system. Provides workflow state management, sub-agent coordination, and session persistence.

Research pipeline for topology-aware GNN representation learning on power grids using the PowerGraph benchmark. Use when (1) building physics-guided GNNs for power flow (PF), optimal power flow (OPF), or cascading failure prediction, (2) implementing self-supervised pretraining for power systems, (3) evaluating cascade explanation fidelity against ground-truth masks, or (4) conducting reproducible ML-for-power-systems research. Triggers include "PowerGraph", "power flow GNN", "OPF surrogate", "cascade prediction", "physics-guided GNN", "grid analytics ML", "power system representation learning".

Perform DuckDuckGo web searches and return structured results. WHEN: User wants to "search the web", "find information online", "look up", "research topic". WHEN NOT: Know exact URL (use browser_navigate), crawling sites (use scrape_url).

Debug Neovim/LazyVim configuration issues. Use when: user reports Neovim errors, keymaps not working, plugins failing, or config problems. Provides systematic diagnosis through hypothesis testing, not just checklists. Think like a detective narrowing down possibilities.

Integrate new DEX aggregators, swappers, or bridge protocols (like Bebop, Portals, Jupiter, 0x, 1inch, etc.) into ShapeShift Web. Activates when user wants to add, integrate, or implement support for a new swapper. Guides through research, implementation, and testing following established patterns. (project)

This skill should be used when you need AI debate with multiple perspectives, analyzing trade-offs, evaluating pros and cons, deep code analysis, or multi-model reasoning. Ideal for subjective decisions, architectural trade-offs, and complex problems requiring synthesis of different viewpoints. Complements opus-subagent for structured multi-step workflows.

Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.

Generate and edit images using Google's Nano Banana Pro (Gemini 3 Pro Image) API. Use when the user asks to generate, create, draw, design, make, edit, modify, change, alter, or update images. Also use for "make me an image", "create a picture", or "draw me a...". Use when user references an existing image file and asks to modify it in any way (e.g., "modify this image", "change the background", "replace X with Y"). Supports both text-to-image generation and image-to-image editing with configurable resolution (1K default, 2K, or 4K for high resolution). DO NOT read the image file first - use this skill directly with the --input-image parameter.

Transform prompts into structured TCRO format with phase-specific clarification. Automatically invoked by /ai-eng/research, /ai-eng/plan, /ai-eng/work, and /ai-eng/specify commands. Use when refining vague prompts, structuring requirements, or enhancing user input quality before execution.

Identify research gaps from systematic literature reviews. Use when: (1) Completing literature reviews, (2) Justifying new studies, (3) Grant proposal development, (4) Dissertation planning, (5) Identifying future research directions.

Research and scaffold demo applications. Use when starting a new demo app, analyzing UI patterns from real apps, or setting up demo app structure.

Validate skill documents against PMC skill design principles.Checks format ownership, composite skill structure, and internal consistency.CHECKS:1. FORMAT OWNERSHIP - kb skill owns all format templates (kb/references/)2. COMPOSITE STRUCTURE - Steps must have associated skills, stage announcements3. INTERNAL CONSISTENCY - No gaps, broken references, or missing sectionsUse when:- Creating new skills- Reviewing existing skills- Periodic skill maintenance- User says "lint skills", "check skills", "validate skills"