MCP Server

Tkeron includes a Model Context Protocol (MCP) server that exposes comprehensive documentation to AI agents and development tools.

What is MCP?

The Model Context Protocol is a standard for providing context to AI agents. Tkeron's MCP server allows AI assistants to access up-to-date documentation about how to use Tkeron.

Installation

The MCP server is included when you install Tkeron globally:

bun install -g tkeron

This installs three commands:

tk / tkeron — The main CLI tool
tkeron-mcp — The MCP server

Configuration

VS Code

Add to ~/.config/Code/User/mcp.json:

{
  "servers": {
    "tkeron": {
      "command": "tkeron-mcp"
    }
  }
}

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "tkeron": {
      "command": "tkeron-mcp"
    }
  }
}

Other Clients

Any MCP-compatible client can run tkeron-mcp. It communicates via stdio transport.

Available Resources

URI	Description
`tkeron://overview`	What Tkeron is and does
`tkeron://getting-started`	Installation and first project
`tkeron://components-html`	HTML components (.com.html)
`tkeron://components-typescript`	TypeScript components (.com.ts)
`tkeron://components-markdown`	Markdown components (.com.md)
`tkeron://pre-rendering`	Pre-rendering (.pre.ts)
`tkeron://cli-reference`	CLI documentation
`tkeron://best-practices`	Patterns and anti-patterns
`tkeron://common-issues`	Troubleshooting guide
`tkeron://testing`	Testing with getBuildResult()

Available Tools

Example Management

list_examples — List all available Tkeron example projects
get_example — Get the complete source code of an example

The MCP server only provides documentation and examples. It does NOT modify files. AI agents use terminal commands (tk init, tk build, tk dev) to manage projects.

What AI Agents Can Do

With access to the MCP server, AI agents can:

✅ Understand Tkeron's capabilities and limitations
✅ Help create components correctly
✅ Suggest appropriate patterns
✅ Provide accurate CLI commands
✅ Debug common issues
✅ List and explore example projects
✅ Initialize, build, and develop projects via terminal

Benefits

For Users

Get accurate help from AI assistants
AI understands your project structure
Contextual suggestions based on docs
Fewer mistakes, faster development

For AI Agents

Access to authoritative documentation
Clear understanding of limitations
Examples and patterns to reference
Ability to distinguish what Tkeron can and can't do

Troubleshooting

"Command not found: tkeron-mcp"

bun install -g tkeron
which tkeron-mcp

MCP Client Not Connecting

Check JSON syntax in config file
Verify correct command path
Restart your IDE/editor

Next Steps

Overview — Learn what Tkeron is
Getting Started — Start using Tkeron
Best Practices — Patterns and anti-patterns