clonetime-mcp-server

MCP (Model Context Protocol) server for Clonetime - the Build vs Buy Calculator.

Analyze SaaS products directly from Claude Desktop, VS Code, or any MCP-compatible AI assistant.

Quick Start

1. Install Claude Desktop

Download from claude.ai/download

2. Configure MCP

Add to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "clonetime": {
      "command": "npx",
      "args": ["clonetime-mcp-server"]
    }
  }
}

3. Restart Claude Desktop

Close and reopen Claude Desktop to load the MCP server.

4. Start Analyzing!

Ask Claude:

• "Analyze Stripe for build vs buy"
• "How long would it take to build a Notion clone?"
• "Compare Slack and Discord complexity"

Available Tools

clonetime_analyze

Analyze a SaaS product to estimate build vs buy complexity.

Parameters:

• input (required): Product URL or name (e.g., "https://stripe.com" or "Stripe")
• apiDocsUrl (optional): URL to API documentation for better analysis
• technicalDocsUrl (optional): URL to technical documentation

Returns:

• MVP estimate (days)
• Production estimate (months)
• Feature breakdown by complexity
• Technical details (architecture, API complexity, integrations)

clonetime_get_analysis

Retrieve an existing analysis by its share token.

Parameters:

• token (required): Share token from a previous analysis

clonetime_compare

Compare 2-4 products side-by-side.

Parameters:

• tokens (required): Array of share tokens to compare

clonetime_list_examples

Get example analyses to see what Clonetime can do.

Parameters:

• count (optional): Number of examples (default: 3, max: 10)

Example Conversation

User: Can you analyze Notion for build vs buy?

Claude: I'll analyze Notion using Clonetime.

[Uses clonetime_analyze tool]

# Notion

**URL:** https://notion.so

## Estimates
- **MVP:** 45-60 days
- **Production-Ready:** 8-12 months

## Summary
Notion is a collaborative workspace platform combining documents, databases,
wikis, and project management. The core challenge is the block-based editor
with real-time collaboration...

## Features by Complexity

### Very High Complexity
- **Real-time Collaboration**: Requires CRDT or OT algorithms...
- **Block-based Editor**: Custom editor with 50+ block types...

### High Complexity
- **Database Views**: Multiple view types (table, kanban, calendar)...
...

Configuration

Environment Variables

| Variable | Description | Default | |----------|-------------|---------| | CLONETIME_API_URL | Base URL for Clonetime API | https://clonetime.ai |

Custom API URL

To use a custom Clonetime instance:

{
  "mcpServers": {
    "clonetime": {
      "command": "npx",
      "args": ["clonetime-mcp-server"],
      "env": {
        "CLONETIME_API_URL": "https://your-instance.com"
      }
    }
  }
}

Rate Limits

The Clonetime API has rate limits:

• 10 analyses per hour per IP address
• Cached analyses can be retrieved without limits

Development

Build from Source

git clone https://gitlab.com/joaocostafernandes-group/clonetime.ai.git
cd clonetime.ai/clonetime/mcp-server
npm install
npm run build

Run Locally

node dist/index.js

Test with MCP Inspector

npx @modelcontextprotocol/inspector ./dist/index.js

The VIBE CODING Philosophy

Clonetime is built on the VIBE CODING philosophy:

"With AI assistance, a skilled developer can now accomplish in days what used to take weeks. But some things can't be accelerated: compliance reviews, security audits, user research, and market validation still take their time."

The estimates account for:

• 10-20x acceleration for code generation with AI
• Realistic timelines for non-code work
• Complexity awareness - some features are inherently harder regardless of AI

Links

• Clonetime Website
• MCP Documentation
• Model Context Protocol
• Report Issues

License

MIT