Model Context Protocol

Connect AI Clients
to Your Workspace

Use Claude, Cursor, VS Code, or any MCP-compatible AI to build apps, query data, and publish websites — all with natural language.

URLhttps://mcp.agentery.ai/mcp|AuthX-API-Key: agt_pub_...

Up and running in 3 steps

01

Get Your API Key

Sign in to Agentery, go to Settings → API Keys, and create a free MCP key.

Open Dashboard
02

Pick Your AI Client

Choose Claude Desktop, Claude Code, Cursor, VS Code, or any MCP-compatible AI tool.

03

Paste Config & Chat

Add the config snippet for your client, restart, and start building with natural language.

Connect your AI client

Replace agt_pub_YOUR_KEY with your key from the dashboard.

🖥️

Claude Desktop

{
  "mcpServers": {
    "agentery": {
      "url": "https://mcp.agentery.ai/mcp",
      "headers": {
        "X-API-Key": "agt_pub_YOUR_KEY"
      }
    }
  }
}

Save to ~/.config/claude-desktop/config.json (macOS/Linux) or %APPDATA%\Claude\config.json (Windows). Restart Claude Desktop.

⌨️

Claude Code

claude mcp add agentery https://mcp.agentery.ai/mcp \
  --header "X-API-Key: agt_pub_YOUR_KEY"

Run once in your terminal. Verify with: claude mcp list

🎯

Cursor

{
  "servers": {
    "agentery": {
      "url": "https://mcp.agentery.ai/mcp",
      "headers": {
        "X-API-Key": "agt_pub_YOUR_KEY"
      }
    }
  }
}

Save to .cursor/mcp.json in your project, or add in Cursor → Settings → MCP Servers.

💻

VS Code / Copilot

{
  "mcp.servers": {
    "agentery": {
      "url": "https://mcp.agentery.ai/mcp",
      "headers": {
        "X-API-Key": "agt_pub_YOUR_KEY"
      }
    }
  }
}

Add to .vscode/settings.json or your VS Code user settings.json.

🌊

Windsurf

{
  "mcpServers": {
    "agentery": {
      "url": "https://mcp.agentery.ai/mcp",
      "apiKey": "agt_pub_YOUR_KEY"
    }
  }
}

Add to Windsurf MCP config. Replace YOUR_KEY with your Agentery API key.

6 public tools — ready to use

Public API keys give access to these tools. No approval needed.

create_app

Create a new app from a natural-language description

e.g. "Create a customer feedback survey"

list_apps

List all apps in your workspace

e.g. "Show me all my apps"

get_app

Get full details + blueprint for any app

e.g. "Show me the NPS survey blueprint"

query_data

Query data rows from any app table

e.g. "Get survey responses where score < 7"

publish_app

Publish an app version live

e.g. "Publish my feedback form"

auto_build_website

AI-generate a complete website end-to-end

e.g. "Build a SaaS marketing site with pricing"

Need QA testing or eval tools? Request a QA key →

See it in action

User → Build me a customer feedback form with NPS and open comments
Claude → [calls create_app] Creating “Customer Feedback Form”...
Claude → App created! App ID: app_abc123. Building now...
User → Get the last 10 responses where NPS score is below 7
Claude → [calls query_data] Found 4 responses. 3 mentioned “slow support”, 1 mentioned “pricing”.
User → Publish it live
Claude → [calls publish_app]✓ Published! Your form is live at app_abc123.agentery.ai

Get help via MCP

Once connected, you can ask your AI client questions about Agentery and get instant answers — without leaving your coding workflow.

You: “How do I add a Google Sheets connector to my app?”
AI: Use list_connectors to see available connectors, then execute_action with action “connect_google_sheets”...

Ready to connect?

Create a free account to get your API key in under a minute.