VitalyGPT MCP
The VitalyGPT MCP server gives you access to Vitaly Friedman's UX insights as a searchable knowledge base. You can use it to:
search_insights: Full-text search across all insights using a keyword or phrase (e.g. 'dropdown', 'mobile navigation', 'usability testing').get_topic: Retrieve all insights for a specific UX topic area using a topic slug (e.g. forms, navigation, research).list_topics: Discover all available UX topic areas, including Forms & Inputs, Navigation & IA, Research & Usability Testing, UX Strategy & Metrics, Design Process, Careers & Communication, UI Components & Patterns, and AI & Design Tools.get_random_insight: Fetch a random UX insight for design inspiration.
Allows OpenAI Codex CLI to search and browse Vitaly Friedman's UX insights for design advice.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@VitalyGPT MCP search for accessible form design insights"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
VitalyGPT MCP
A Model Context Protocol (MCP) server that makes Vitaly Friedman's UX insights available as a tool to any LLM.
Vitaly is the founder of Smashing Magazine and one of the most prolific UX educators on the internet. This project collects his insights and serves them as structured, searchable knowledge — so you can reference his thinking directly inside Claude, Cursor, or any MCP-compatible client while designing your next product.
Status: Personal use / early preview. Shared with Vitaly for feedback before wider distribution.
What it does
Four tools, zero setup friction:
Tool | Description |
| See all UX topic areas covered |
| Full-text search across all insights |
| Get all insights for a specific topic |
| Random insight for design inspiration |
Topics covered: Forms & Inputs · Navigation & IA · Research & Usability Testing · UX Strategy & Metrics · Design Process · Careers & Communication · UI Components & Patterns · AI & Design Tools
Related MCP server: systembridge-mcp
Setup
Requirements: Node.js 18+
Fastest way: ask your AI agent
Paste this into whatever AI coding agent you use — Claude Code, Cursor, Codex, Gemini, all of them:
Read the setup instructions at github.com/setugk/vitalygpt and add its MCP server (
npx -y @vitalygpt/mcp) to your own config in whatever way is correct for you.
It'll fetch this page, figure out the right config mechanism for itself, and tell you when to restart.
Manual setup
Prefer to do it yourself? All clients use the same server command — only the config file's location (and sometimes its format) differs. After editing, restart the client so it picks up the change.
Claude Code
Create .mcp.json at the root of the folder/workspace you open in Claude Code (not a subfolder — it won't be discovered otherwise):
{
"mcpServers": {
"vitalygpt": {
"command": "npx",
"args": ["-y", "@vitalygpt/mcp"]
}
}
}Restart Claude Code, then confirm with /mcp or by asking "what MCP servers are connected?".
Claude Desktop
Edit claude_desktop_config.json:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.json
Same JSON as above (top-level mcpServers key). Restart the app.
Cursor
Create .cursor/mcp.json — either in a project root (project-only) or ~/.cursor/mcp.json (all projects):
{
"mcpServers": {
"vitalygpt": {
"command": "npx",
"args": ["-y", "@vitalygpt/mcp"]
}
}
}OpenAI Codex CLI
Add to ~/.codex/config.toml (or .codex/config.toml in a trusted project):
[mcp_servers.vitalygpt]
command = "npx"
args = ["-y", "@vitalygpt/mcp"]Or via the CLI: codex mcp add vitalygpt -- npx -y @vitalygpt/mcp
Gemini CLI
Add to ~/.gemini/settings.json (global) or .gemini/settings.json (project):
{
"mcpServers": {
"vitalygpt": {
"command": "npx",
"args": ["-y", "@vitalygpt/mcp"]
}
}
}Contributing
Content lives in content/ as plain markdown files — one file per topic. Each insight follows this format:
## Insight Title
One-paragraph summary of the key takeaway.
**Full post:**
The original post content.
**Source:** [LinkedIn](https://linkedin.com/...) · Vitaly Friedman · YYYY-MM-DD
**Tags:** tag1, tag2, tag3To add a new insight: edit the relevant topic file and submit a PR. No code knowledge required.
Topic files:
content/forms-and-inputs.mdcontent/navigation-and-ia.mdcontent/research-and-testing.mdcontent/strategy-and-metrics.mdcontent/process-and-workflow.mdcontent/careers-and-communication.mdcontent/components-and-patterns.mdcontent/ai-and-tools.md
Roadmap
Publish to npm as
@vitalygpt/mcpDeduplicate near-identical entries
Add semantic search (embeddings)
Support for contributions from Vitaly directly
Attribution
All insights are sourced from Vitaly Friedman's public LinkedIn posts, with attribution and links to originals. This project is not affiliated with Vitaly or Smashing Magazine.
License
MIT
Available Tools
4 toolsget_random_insightA
Get a random UX insight from Vitaly Friedman. Great for inspiration during design sessions.
Inputs: no named parameters.
get_topicA
Get all UX insights for a specific topic area (e.g. forms, navigation, research).
Inputs: topic (required).
list_topicsA
List all available UX topic areas covered in the knowledge base.
Inputs: no named parameters.
search_insightsA
Search Vitaly Friedman's UX insights by keyword or topic. Returns matching insights with full content.
Inputs: query (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool has a clearly distinct purpose: random insight, topic-specific insights, topic listing, and keyword search. No overlap or ambiguity.
All tool names follow a consistent verb_noun pattern in snake_case (get_random_insight, get_topic, list_topics, search_insights).
With 4 tools, the server is well-scoped for a UX insight knowledge base, covering random retrieval, topic browsing, and search without excess.
The tools cover the core needs: random inspiration, topic exploration, topic listing, and full-text search. No obvious gaps for the given domain.
Maintenance
Related MCP Connectors
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
The Remote MCP server acts as a standardized bridge between LLM applications (like Claude, ChatGPT, and Cursor) and external services, enabling AI agents to access external tools and resources. Its primary capability is providing a centralized search tool to discover other MCP servers and their respective tools. Unlike local implementations, it runs remotely with OAuth authentication and permission controls for security.
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
An MCP memory server. One memory your agents share — across models, devices and apps.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceMCP tool server that gives any AI agent the ability to search, scrape, and analyze content across the internet.42MIT
- AlicenseAqualityDmaintenanceMCP server that gives LLMs deep knowledge of design systems and tokens, enabling intelligent design evolution, token analysis, and designer-to-developer handoffs.3718 npm2MIT
- AlicenseNot gradedqualityDmaintenanceA remote MCP server exposing uxspot.io's curated UX glossary, AI tools directory, checklists, and Learn UX curriculum as callable tools. Enables AI assistants to search for UX terms, tools, and checklists via natural language.2MIT
- AlicenseNot gradedqualityDmaintenanceMCP server for design decisions during a build, providing live UI/animation library docs (via llms.txt) and a curated knowledge base of UX laws and UI conventions. It enables agents to look up real component APIs and apply design best practices, running locally via stdio or hosted on Vercel.6 npmMIT