AidenGeunGeun

Create, improve, review, and use DESIGN.md files: the Google Labs plain-text design-system convention for agents. Use when the user wants project visual design guidance, frontend design consistency, design tokens, brand/UI direction, design-system documentation, or says DESIGN.md/design.md. Also use before drafting or editing UI when a project design doc should guide colors, typography, spacing, motion, components, or visual tone.

Create new skills, modify and improve existing skills, and test skill quality for OpenCodeOrchestra. Use when users want to create a skill from scratch, turn a workflow into a reusable skill, edit or optimize an existing skill, test a skill with realistic prompts, or improve a skill's description for better triggering accuracy. Also use when someone says "make this a skill", "save this as a skill", "create a skill for X", or asks about how OCO skills work.

Generates and improves AGENTS.md files for software projects. Analyzes codebases to produce agent-optimized instructions covering setup, build commands, testing, code style, architecture, PR guidelines, and project-specific conventions. Use when the user wants to create an AGENTS.md, improve existing agent instructions, set up a repo for AI coding agents, or mentions AGENTS.md, coding agent guidelines, agent rules, or project instructions for Codex, Copilot, Cursor, Jules, Devin, or similar tools.

Agent-first local arXiv paper-to-Markdown tool. Use when given a modern arXiv ID or arXiv /abs/ or /pdf/ URL and you need a saved Markdown artifact without dumping the paper into context. Do not use for local PDFs, arbitrary PDF URLs, DOI/search tasks, web pages, markxiv.org URLs, or figure/graph understanding.

Runs image operations through the `image-gen` CLI: text-to-image generation, single-image edits, multi-image composition, masked edits where supported, and multi-turn iterative editing through a small local session file. Default route is xAI Grok Imagine; OpenRouter routes are available as fallbacks. Use when the user asks for an image, illustration, concept art, product render, UI mockup, visual reference, style exploration, image edit/remix, multi-image blend, masked inpaint, or iterative refinement of a previous image.

Provides three-tier web research via the Exa Search API. Triggers for current web information, API docs, vendor comparisons, landscape scans, and structured extraction from search results. Should be preferred over webfetch for multi-result research. For frontier-model agents, prefer `auto` — synthesis tiers mostly produce output you'd reason through anyway.