Component Harvester MCP Server
Harvests React components from Aceternity UI and Shadcn UI, fetching code and metadata to generate integration prompts for AI agents.
README
Component Harvester MCP Server β¨
This MCP (Model Context Protocol) server is your AI agent's trusty sidekick for sourcing React components from popular UI libraries. It automates finding, fetching, and understanding how to integrate UI elements into your projects, supporting both direct code harvesting and CLI-based integration guidance.
Supported Libraries
- Aceternity UI: Full component code and metadata harvesting.
- Shadcn UI: Component metadata harvesting and CLI-based integration guidance.
- (More to come soon!)
TLDR;
This server fetches component information from Aceternity UI and Shadcn UI.
- For Aceternity UI, it gets the full code and dependencies.
- For Shadcn UI, it gets metadata (like dependencies, files created) and tells you how to install the component using its CLI.
The server stores this information and can generate detailed prompts for an AI agent (or you!) to integrate these components.
Example Agent Prompts:
"Fetch the '3D Pin' component from Aceternity UI and show me the code and dependencies.""List all available components from Aceternity UI.""Scan the 'Alert Dialog' component from Shadcn UI and show me how to integrate it.""List all available Shadcn UI components."
What Can It Do? π
The Component Harvester enables an AI agent to:
- Discover Components: Automatically learn about available components by fetching and parsing the main JSON registries from Aceternity UI (
https://ui.aceternity.com/registry) and Shadcn UI (https://ui.shadcn.com/registry). - Fetch Component Data:
- Aceternity UI: Grabs detailed, structured JSON data for specific components (e.g., from
https://ui.aceternity.com/registry/[slug].json), which includes full source code, file paths, dependencies, etc. - Shadcn UI: Retrieves metadata for specific components directly from its main registry JSON. This includes dependencies, file names, etc., but not the full source code (as Shadcn components are added via CLI).
- Aceternity UI: Grabs detailed, structured JSON data for specific components (e.g., from
- Store Locally:
- Maintains a central index (
data/harvested_index.json) of all processed components from all sources. - Stores detailed data for each component in source-specific directories:
data/aceternity/[slug].json: Contains the full fetched JSON (including code) for an Aceternity UI component.data/shadcn/[slug].json: Stores the metadata (name, type, deps, file list) for a Shadcn UI component, as extracted from their main registry.
- Maintains a central index (
- Generate Integration Blueprints: Produces detailed, step-by-step prompts:
- Aceternity UI Prompts: Include full code, file paths, and dependency information for manual integration.
- Shadcn UI Prompts: Guide using
npx shadcn-ui@latest add [slug], list expected files, and dependencies.
How It Works π§ββοΈ
1. Learning the Lay of the Land: Registry Processing
When the server starts, it fetches the main registry JSON directly from both Aceternity UI and Shadcn UI.
graph TD
subgraph Server Startup
direction LR
A1[Initiate] --> A2a[Fetch JSON from aceternity.com/registry]
A2a --> A3a[Parse Aceternity JSON & Cache Slugs]
A1 --> A2b[Fetch JSON from shadcn.com/registry]
A2b --> A3b[Parse Shadcn JSON & Cache Metadata]
end
A3a --> A4[Server Ready]
A3b --> A4
- The Goal: To build internal maps (
aceternityRegistryDataandshadcnRegistryData) for quick lookups.- For Aceternity, it maps a normalized component name to its slug (e.g., "3D Pin" -> "3d-pin").
- For Shadcn, it maps a normalized component name to its full metadata object from the registry.
2. The Treasure Hunt: Fetching/Processing Specific Component Data
Aceternity UI: When an agent requests an Aceternity component (e.g., "3D Pin"), the server uses its cached slug.
graph TD
B0[Agent Request: scan_aceternity_component] --> B1[MCP Server]
B1 --> B2{Slug in Aceternity Cache?}
B2 -- Yes --> B3[Construct URL: aceternity.com/registry/slug.json]
B3 --> B4[Fetch Component JSON via Axios]
B4 -- Full JSON Data --> B5[Parse & Store Full Component Data]
B5 --> B6[Update Main Index & In-Memory Cache]
B6 --> B7[Report Success]
B2 -- No --> B8[Report Not Found]
- Data Fetched: The complete JSON for the component, including source code for all its files.
Shadcn UI: When an agent requests a Shadcn component (e.g., "Alert Dialog"):
graph TD
C0[Agent Request: scan_shadcn_component] --> C1[MCP Server]
C1 --> C2{Component in Shadcn Cache?}
C2 -- Yes --> C3[Retrieve Metadata from Cache]
C3 -- Metadata (files, deps) --> C4[Store This Metadata as Component's Data]
C4 --> C5[Update Main Index & In-Memory Cache]
C5 --> C6[Report Success]
C2 -- No --> C7[Report Not Found]
- Data Processed: The metadata (dependencies, file list, etc.) already fetched from
https://ui.shadcn.com/registry. No individual[slug].jsonis fetched for Shadcn as it's not provided by them.
3. Keeping a Record: Component Storage
All processed component data is stored:
- Individual JSON Files:
data/aceternity/[slug].json: Stores the full JSON (code, files, deps) for an Aceternity UI component.data/shadcn/[slug].json: Stores the metadata (name, type, deps, file list) for a Shadcn UI component, as extracted from their main registry.
- Central Index File (
data/harvested_index.json): A manifest of all components from all sources, with metadata like name, source, slug, description, path to its specific JSON file, and last scan time.{ "aceternity:3DPin": { /* ... metadata ... */ }, "shadcn:AlertDialog": { /* ... metadata ... */ } }
An in-memory cache (inMemoryIndexCache) mirrors this index for fast access.
4. The Blueprint: Generating Integration Prompts
The get_aceternity_component_prompt and get_shadcn_component_prompt tools transform stored data into integration instructions.
- Aceternity UI Prompts: Include full code, file paths, and dependency information.
- Shadcn UI Prompts: Guide using
npx shadcn-ui@latest add [slug], list expected files, and dependencies.
Meet the Tools: MCP Server Endpoints π οΈ
Common Tools
list_harvested_components: Lists all components stored in the index.- Input:
{ "source": "aceternity" | "shadcn" | "all" }(optional)
- Input:
Aceternity UI Tools
scan_aceternity_component: Fetches and stores full data for an Aceternity UI component.- Input:
{ "componentName": "string" }
- Input:
get_aceternity_component_prompt: Generates a detailed integration guide (with code) for a scanned Aceternity UI component.- Input:
{ "componentName": "string" }
- Input:
Shadcn UI Tools
list_shadcn_components: Lists all components available in the Shadcn UI registry cache.- Input:
{}
- Input:
scan_shadcn_component: Processes and stores metadata for a Shadcn UI component from its main registry.- Input:
{ "componentName": "string" }
- Input:
get_shadcn_component_prompt: Generates an integration guide (CLI-focused) for a Shadcn UI component.- Input:
{ "componentName": "string" }
- Input:
The Road Ahead: Future Enhancements πΊοΈ
- Smarter Prompt Generation: More context-aware advice in prompts.
- Broader Horizons: Adding support for more component libraries.
- Manual Cache Refresh: Tools to re-trigger registry fetching on demand.
- Enhanced Error Handling: Continuous improvements for robustness.
We're excited about making component integration smoother and faster!
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.