@vampsieee/techterrain-mcp

An MCP server that lets AI agents explore 3.4 million years of technology history through TechTerrain — a 3D voxel timeline of 1,200+ curated milestones.

Capture screenshots, generate GIFs, search any topic, and analyze cross-domain patterns — all from Claude, OpenClaw, or any MCP-compatible client.

Install

npx -y @vampsieee/techterrain-mcp

No API key required. The server connects to the public TechTerrain API at techterrain.io.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "techterrain": {
      "command": "npx",
      "args": ["-y", "@vampsieee/techterrain-mcp"]
    }
  }
}

Restart Claude Desktop. You'll see the hammer icon with 7 tools.

Try It

"Explore quantum computing on TechTerrain and take a screenshot"

"Show me how AI and biotech have co-evolved, then generate a GIF orbiting the scene"

"What causal relationships exist between technology categories?"

Tools

| Tool | Description | Cost | |------|-------------|------| | capture_screenshot | Headless screenshot via Cloudflare Browser Rendering → CDN URL | Free (35/hr) | | capture_gif | Animated GIF generation (orbit or timelapse) → async job | Free (35/hr) | | get_gif_status | Poll async GIF job progress and get result URL | Free | | explore_topic | Search any topic → mapped milestones with years, categories, impact | LLM cost | | generate_insights | LLM pattern analysis of a scene descriptor | LLM cost | | list_captures | Browse the Agent Showcase gallery | Free | | get_insights | Pre-computed Granger causality findings | Free |

Resources

| Resource | URI | Description | |----------|-----|-------------| | Agent API Docs | techterrain://docs | Full API reference with code examples | | LLM Documentation | techterrain://llms.txt | Concise API reference for LLMs | | Dataset | techterrain://dataset | Raw milestone dataset (1,200+ entries) |

Example Workflow

User: "Show me how AI and biotech have co-evolved"

Agent:
1. explore_topic("AI biotech co-evolution")
   → Gets 15 milestones mapped across both categories
2. capture_screenshot({
     injectJs: "window.__techterrain.navigate({goto: 2010, categories: ['AI', 'Health']})",
     caption: "AI-Biotech convergence 2010s"
   })
   → Gets CDN URL of the rendered view
3. get_insights({ limit: 5 })
   → Gets pre-computed causal findings about AI↔Health correlations
4. Returns: image, milestones, and statistical insights to user

Environment Variables

| Variable | Default | Description | |----------|---------|-------------| | TECHTERRAIN_URL | https://techterrain.io | Base URL of the TechTerrain instance |

Development

cd mcp
pnpm install
npm run dev      # Run with tsx (hot reload)
npm run build    # Build distributable

License

MIT