"Semantic UI" matching MCP servers:

Enables natural language search of local photo archives using AI-powered semantic understanding, with integration into Claude Desktop via the Model Context Protocol.

Delegates frontend/UI refinement to Gemini via agy CLI, using iterative vision loops and diff-gating to realign UIs to design mockups without affecting backend code.

An MCP server that gives models semantic context about a data warehouse, including table meaning, units, lineage, and health, to reduce guessing.

MCP file cache for AI coding agents. Re-reading an unchanged file costs ~40 tokens instead of the whole file; a changed file comes back as a diff. Adds grep, BM25 search and glob over cached files, atomic edits with dry_run, and token-budgeted batch reads. One local daemon serves every client. No API keys, no GPU. Works with Claude Code, Cursor, VS Code, Codex and OpenCode.

MCP server that lets AI agents search and inspect shadcn components from the encode-ui registry, returning metadata, source code, and install commands.

Visual CVE audit dashboard for npm, Python, Go, and Rust projects. Scans your project manifests (package-lock.json, requirements.txt, go.sum, Cargo.lock) against OSV.dev live data, opens a browser dashboard for human review, then applies fixes only after explicit confirmation. Supports multi-service monorepos in one command.

A local MCP server that provides semantic code search for Python codebases using tree-sitter for chunking and LanceDB for vector storage. It enables natural language queries to find relevant code snippets based on meaning rather than just text matching.

Enables auto-configuration and validation of React components against a glassmorphic design token system using an AST correction engine. It supports managing UI presets, enforcing design uniformity in Monaco IDE, and exporting tokens as CSS, Tailwind, or JavaScript variables.

Query Apple Health export data with an LLM via a semantic layer that avoids common data traps, enabling accurate natural-language queries about health metrics. Optionally imports CPAP therapy data for correlation.

MCP server for see-u-ui component library, allowing AI assistants to query component props/events/slots/examples on demand.

An MCP server that lets AI agents see, act on, and verify the rendered Neovim UI, like a Playwright for Neovim. It provides tools to launch/attach to Neovim, observe the screen, send input, and wait for conditions.

Bidirectional MCP bridge enabling AI assistants to draw UI on Figma canvas and read designs back as structured data or code tokens via localhost, requiring Figma Desktop.

A local stdio MCP server that exposes compact semantic observations from UI elements annotated with data-agent-* hints, designed to work alongside the official Playwright MCP for browser automation.

Provides AI coding assistants with on-demand access to component specs, test scenarios, accessibility requirements, and build guides from the Web UI Component Specification.

Enables MCP clients to read design structure, take screenshots, create nodes, and edit UI directly on Figma canvas via a bridge between MCP and Figma Desktop.

MCP server for the coss ui design system, enabling AI agents to fetch components, props, and design tokens on demand for accurate, low-cost code generation.

MCP server that helps AI agents create UIs that don't look AI-generated by auditing components, refining copy, suggesting themes, and checking accessibility/contrast.

Local semantic recall MCP server for citation-integrity workflows. Enables semantic search over a text corpus, returning file paths and page numbers for verification, with 100% local processing using ONNX embeddings and LanceDB.

MCP server that annotates any web page with hover labels — zero extensions, works in any browser

Enables coding agents to see the UI they just changed without flooding their context window, returning a structured text summary (console errors, failed requests, overflow, layout shifts, accessibility violations) plus small crops of only the regions that changed since the last capture. All output is fitted into a chosen token budget, with captures stored on disk and evicted automatically.