heroui-v2-mcp
A standalone local MCP server that serves HeroUI v2 component documentation, source code, style definitions, and theme tokens by fetching directly from the official GitHub repository.
README
heroui-v2-mcp
MCP (Model Context Protocol) server for HeroUI v2 (https://v2.heroui.com) component documentation, source code, style definitions and theme tokens.
Unlike the official @heroui/react-mcp — which targets HeroUI v3 and needs a
deployed Cloudflare Worker + R2 bucket + PostHog project behind it — this is a standalone local server:
it fetches everything on demand directly from heroui-inc/heroui@main (the branch that tracks v2.x) and
raw.githubusercontent.com, caching results in memory for the life of the process. No cloud account,
API deployment or extraction pipeline required.
Why a separate project
HeroUI v3 (branch v3, site heroui.com) and v2 (branch main, site v2.heroui.com) are different,
incompatible component models — different package layout, no llms.txt on the v2 site, different docs
folder structure (/docs/components/* vs /docs/react/*), different styling approach (tailwind-variants
files under packages/core/theme/src/components vs whatever v3 uses). The official MCP explicitly refuses
to serve v2 docs, so this project mirrors its tool surface but points at the v2 data sources.
Tools
| Tool | Purpose |
|---|---|
list_components |
List all v2 component slugs (matches v2.heroui.com/docs/components/<slug>) |
get_component_docs |
Description, install command, imports, usage guidance, props/API table |
get_component_source_code |
React/TypeScript source from packages/components/<pkg>/src |
get_component_source_styles |
tailwind-variants style definition from packages/core/theme/src/components |
get_docs |
General guides: installation, frameworks, theming, CLI, HeroUIProvider, etc. |
get_theme_variables |
Default semantic colors + layout tokens |
Install & run
The package is published on npm as heroui-v2-mcp, so no
clone or build step is required to use it.
Use from Claude Code / Claude Desktop / Cursor
Add to your MCP config — npx downloads and runs it on demand:
{
"mcpServers": {
"heroui-v2": {
"command": "npx",
"args": ["-y", "heroui-v2-mcp"]
}
}
}
Or install it globally and point at the binary directly:
npm install -g heroui-v2-mcp
{
"mcpServers": {
"heroui-v2": {
"command": "heroui-v2-mcp"
}
}
}
Local development (from source)
npm install
npm run dev # tsx --watch, runs src/server.ts directly (no build step)
npm run inspect # launches @modelcontextprotocol/inspector against src/server.ts
npm run build # emits dist/server.js
GitHub rate limits
Directory listings go through the GitHub REST API, which is capped at 60 requests/hour
unauthenticated. File contents (raw.githubusercontent.com) are not subject to that limit. If you hit
403s while browsing many components, set a token:
export GITHUB_TOKEN=ghp_xxx # no scopes needed, just raises the rate limit to 5000/hr
How data is resolved
- Component list = files under
apps/docs/content/docs/components/*.mdxon themainbranch. - Most doc slugs map 1:1 to a
packages/components/<slug>package. A handful of doc pages document sub-components bundled into another package (e.g.textareaships inside@heroui/input,checkbox-groupinside@heroui/checkbox) — seeCOMPONENT_PACKAGE_ALIASESinsrc/constants.ts. - Props/API tables are parsed out of the
<APITable data={[...]} />JSX literal embedded in each.mdxdoc page (that array is evaluated as a JS literal — it's static data from the official repo, not arbitrary/untrusted input). - Theme tokens come from
packages/core/theme/src/colors/{semantic,common}.tsanddefault-layout.ts.
Caching
Every GitHub response (file content or directory listing) is cached in-memory for 30 minutes
(CACHE_TTL_MS in src/constants.ts). Restart the process to force a refresh sooner.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.