io.github.TKMD/reftrixmcpsearch

reftrixmcp

Verified · 22 days ago

Web design analysis with 26 MCP tools: layout, motion, quality, semantic search via pgvector

Install

claude mcp add reftrixmcp -- npx -y @reftrixmcp/[email protected]

Requires: DATABASE_URL, REDIS_URL, OLLAMA_BASE_URL

Our take

This server appears to bundle website analysis tools for layout, motion, semantic search, and general quality checks. It is most relevant to design engineers or teams auditing web experiences rather than general-purpose MCP users. The scope is coherent but very early by visible signals, so it reads as an experimental utility that likely needs hands-on evaluation before depending on it.

reviewed by hand · 2026-07-08

Something wrong or dead here? Report it

Verification record

last verified
22 days ago
github stars
2
last commit
28 days ago
archived
no
license
AGPL-3.0
downloads / wk
31
latest version
0.6.0
in registry since
2026-03-21

github.com/TKMD/ReftrixMCP

mcp-server-searchsearch

mcp-server-search

Verified · 21 days ago

MCP server for web search operations

hand-reviewed39k starschecked 21 days ago

repomixsearch

Repomix

Verified · 12 days ago

Pack local or remote codebases into AI-friendly files that LLMs and coding agents can read or search

hand-reviewed28k stars67k dl/wkchecked 12 days ago

screenpipe-mcpsearch

screenpipe

Verified · 14 days ago

Search your local screen recordings, audio transcripts, and computer activity from screenpipe.

hand-reviewed22k stars2.8k dl/wkchecked 14 days ago

sui-kapasearch

Sui Knowledge Base (kapa.ai)

Verified · 8 days ago

Semantic search across Sui documentation and knowledge sources, powered by kapa.ai

hand-reviewed7.8k starschecked 8 days ago

firecrawl-mcp-serversearch

Firecrawl MCP Server

Verified · 28 days ago

MCP server for Firecrawl — web search, scraping, and biomedical/arXiv paper search.

hand-reviewed7.4k starschecked 28 days ago

robinsearch

Robin: AI-Powered Dark Web OSINT

Verified · 8 days ago

Dark web OSINT over Tor: search onion engines, scrape pages, report with your own model.

hand-reviewed7.3k starschecked 8 days ago