KoliBri MCP API
Model Context Protocol Backend for KoliBri Examples
🟢 Online📡 REST API Endpoints
This API provides structured information about KoliBri samples that can be used by AI agents.
/mcp/samples
List of all available samples. Add the
?q=<search>
query parameter to perform a fuzzy search across component names, tags, titles, descriptions and the component folder path. The search is tolerant
to typos and partial matches, so even short tokens can surface the relevant sample.
The example above demonstrates fuzzy matching—the short token
btn still returns button samples.
Combine multiple words with spaces to narrow the result set—this request filters for responsive table samples.
Path and source code of a specific sample
/mcp/docs
List of Markdown-based documentation documents
Retrieve a specific documentation document including its Markdown source
Retrieve a specific documentation document including its Markdown source (with .md extension)
Note: The indices are generated during the build process. Manual refresh is not available in deployments.
🔧 VS Code Integration
Use the KoliBri MCP Server with GitHub Copilot Chat in VS Code:
Option 1: Local Installation
npm install -g @public-ui/mcp
Install the MCP package globally and create an mcp.json:
{ "servers": { "kolibri-mcp": { "command": "npx", "args": ["@public-ui/mcp"] } }, "inputs": [] }Option 2: Remote Server (Recommended)
Use the hosted server without local installation in your mcp.json:
{ "servers": { "kolibri-mcp": { "url": "https://public-ui-kolibri-mcp.vercel.app/mcp/", "type": "http" } }, "inputs": [] }Setup
Create an mcp.json file in your project directory or use the global configuration:
# Local file in project
echo '{ "servers": { ... } }' > mcp.json
# Or globally for all projects
~/.config/mcp/mcp.json
In GitHub Copilot Chat you can now write:
@kolibri show me a button sample
@kolibri how do I implement a KoliBri table?
@kolibri create an accessible form