An AI skill that provides design intelligence for building professional UI/UX across multiple platforms.
Claude Code Toolbox
Claude Code Toolbox — automated installers and environment configuration framework for Claude Code with one-line setup across Windows, macOS, and Linux.
Install
uvx cc-toolbox setup --profile my-env-2claude mcp add claude-code-toolbox -- uvx cc-toolbox setup --profile my-env-2{
"mcpServers": {
"claude-code-toolbox": {
"command": "uvx",
"args": [
"cc-toolbox",
"setup",
"--profile",
"my-env-2"
]
}
}
}Install command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →Automated installers and an environment configuration framework for Claude Code on Windows, macOS, and Linux.
Define your complete Claude Code environment in a single YAML file -- custom agents, MCP servers, slash commands, hooks, skills, settings, and more -- and install everything with one command.
Features
- Custom agents -- specialized subagents for code review, research, debugging, and any workflow you design
- MCP servers -- HTTP, SSE, and stdio transports with scope-based registration; unchanged servers are skipped on reruns, preserving their OAuth authentication
- Slash commands -- custom commands for frequently used workflows
- Rules -- user-scope rule files for coding standards, security policies, and project conventions
- Skills -- multi-file skill packages for complex agent capabilities
- System prompts -- replace or append to the default Claude Code prompt in the commands an isolated profile installs
- Hooks -- five hook types: command (shell scripts, shell or exec form), HTTP (webhooks), prompt (LLM evaluation), agent (subagent with tools), and MCP tool (a tool on a configured MCP server), plus shared helper modules delivered beside the hook scripts
- Ready-to-run Python hooks -- Python hooks and status-line scripts run through
uv run, and for each one that lists its packages in PEP 723 inline metadata the setup builds its uv environment right after the scripts land, on the same path and with the environment variables your sessions hand the hook, so the first launch inside Claude Code installs nothing; a package that cannot install fails the setup with the script's name and the command to retry, instead of failing the hook within its timeout in a session - User and global settings --
user-settingsis rawsettings.jsoncontent (camelCase keys) andglobal-configis raw~/.claude.jsoncontent, covering model selection, permissions, effort levels, thinking mode, environment variables, and every other Claude Code setting - Status line -- custom status bar scripts for real-time session information
- Configuration inheritance -- extend and override parent configurations with selective per-key merge
- Component selection -- author-defined component groups the end user picks at setup time, interactively (checkbox picker) or via
--select/--with/--without; unclaimed items stay mandatory - Isolated profiles --
command-namesinstalls a configuration as a separate profile under~/.claude/{cmd}/with its own commands, settings,.claude.json, MCP registrations, and environment loaders; a session of the profile reads nothing from the base profile's~/.claudewherever you start it, in your home folder included, because the profile'sconfig.jsonexcludes the baseCLAUDE.mdand rules in every letter case the folder can be spelled in and, on Windows, under the 8.3 short name of your home folder (the spelling of%TEMP%), and the launcher keeps the basesettings.jsonout of a session started in the home folder (that session also loads noCLAUDE.md,CLAUDE.local.mdor.mcp.jsonof the home folder itself; below the home they load as before) and, on Windows, starts the session in the long spelling of the folder you started in -- wherever that folder is, below your home or outside it, with each 8.3 component replaced by the name on disk and the drive letter returned uppercase (c:\workstarts inC:\work) -- so a path spelled by 8.3 short names (the spe
Similar plugs
The 1st Collaborative AI Design Agent Workspace. Local-first and open source. Create prototypes, websites, apps, dashboards, presentations, images, and videos with your team…
Clone any website with one command using AI coding agents
Unofficial Python API and agentic skill for Google Gemini Notebook. Full programmatic access to Gemini Notebook's features—including capabilities the web UI doesn't expose—via…
ARIS (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework,…
Eigent: The Open Source Cowork Desktop - Local and Free Alternative to Claude Cowork and Codex