No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by referodesign · MCP Server · ★ 243
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is refero_skill safe to install? View the security audit →
Refero Skill An agent skill that makes design research mandatory before implementation: styles-first visual research, real-screen pattern research, flow reasoning, reference locks, optional visual exploration, post-build QA, and craft guidance. When Refero MCP is available, it can use curated visual styles, 150,000+ real app screens, and 6,000+ user flows from Stripe, Linear, Notion, Figma, and the best-designed products ever built. Install Works with Claude Code, Cursor, Gemini CLI, Lovable, and any MCP-compatible agent. Craft knowledge loads immediately. No account required. Manual installation On Claude.ai, add the contents of to your project knowledge. What it does Researches styles, screens, and flows — starts with visual styles for taste and direction, then uses real screens and user flows for product patterns and journey logic when live Refero MCP tools are available. Extracts patterns — identifies specific design decisions and builds a reference list before touching code. Routes the workflow — goes directly to code for clear edits, or creates three reference-locked directions when exploration is valuable. Applies craft knowledge — u
| Stars | 243 |
| Forks | 21 |
| Language | Python |
| Category | MCP Server |
| License | MIT |
| Quality Score | 71.1634442178188/100 |
| Open Issues | 2 |
| Last Updated | 2026-09-05 |
| Created | 2026-02-05 |
| Platforms | claude-code, cli, codex, gemini, mcp, python |
| Est. Tokens | ~14k |
These tools work well together with refero_skill for enhanced workflows:
Looking for a refero_skill alternative? If you're comparing refero_skill with other mcp server tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Curated list of awesome AI tools to build beautiful UI/UX.
Official plugin distribution for the hosted Zapier MCP server. Install it in your AI client and connect to tho
Use the official MongoDB Skills with your favorite coding agent to build faster.
Production-ready AI agent profiles for Claude Code, Cursor, Codex, OpenClaw, and Hermes. Search, install, and
A Model Context Protocol (MCP) that allows Claude Desktop and other AI tools (Claude Code, Cursor, Antigravity
Multi-agent orchestration platform for Gemini CLI, Claude Code, Codex, and Qwen Code — 39 specialists, paralle
Explore other popular mcp server tools:
refero_skill is Research-first design skill for AI agents. 150K+ real app screens and flows via Refero MCP.. It is categorized as a MCP Server with 243 GitHub stars.
refero_skill is primarily written in Python. It covers topics such as agent, agent-skills, claude.
You can find installation instructions and usage details in the refero_skill GitHub repository at github.com/referodesign/refero_skill. The project has 243 stars and 21 forks, indicating an active community.
refero_skill is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to refero_skill on Agent Skills Hub include awesome-ai-tools-for-ui, zapier-mcp, agent-skills. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: