Run EasyDeploy from your AI assistant

1. Open Claude in the browser (claude.ai) or open the Claude Desktop app.
2. Open Settings, then Connectors, and choose Add custom connector.
3. Enter exactly:
   • Name: EasyDeploy AI
   • Remote MCP server URL: https://mcp.easydeploy.ai/mcp
4. Save the connector so EasyDeploy appears in your list of connectors.
5. Open the EasyDeploy connector entry and choose Connect, then finish sign-in in your browser. That step authorizes Claude to use your EasyDeploy account.

Claude Desktop: allow network egress

Needed before file uploads and some tool calls work on Desktop

File uploads and some tool calls reach EasyDeploy over the network. On Desktop, Claude blocks outbound traffic unless you explicitly allow the domains it should call.

• Open Settings, then Capabilities.
• Turn on Allow network egress.
• Under the domain allowlist, add this entry exactly (including the leading *.): *.execute-api.us-east-1.amazonaws.com

The domain allowlist UI is available on paid Claude plans.

Claude skills

Optional bundles that work alongside the connector

Each download is a .skill file you upload once in Claude. They are specific to Claude, and the connector works without them.

Data preparation playbook

A detailed data preparation playbook: cleaning tabular data, EDA, leakage checks, and label files ready for EasyDeploy AutoML training. Use with the MCP connector for projects, uploads, training, and predictions.

Download easydeploy.skill

Invoice collection scoring

ML-powered accounts-receivable scoring: collection probability per open invoice, a rescue-first call list, cash flow forecast charts (base, naive, and ML scenarios), and an AR Intelligence dashboard. Requires QuickBooks and the EasyDeploy MCP connector.

Download easydeploy-invoice-score.skill

Add the skill in Claude

1. Open Customize.
2. Open Skills.
3. Choose + Create Skill, then Upload a skill.
4. Upload the .skill file you downloaded (drag and drop or browse). Repeat for each bundle you want. Each file must include SKILL.md inside the bundle; ours do.

One command, with an API key in a header.

claude mcp add --transport http easydeploy https://mcp.easydeploy.ai/mcp \
  --header "Authorization: Bearer eda_live_your_key_here"

Add the server to ~/.cursor/mcp.json. Keep the key in your environment rather than in the file.

{
  "mcpServers": {
    "easydeploy-ai": {
      "url": "https://mcp.easydeploy.ai/mcp",
      "headers": { "Authorization": "Bearer ${env:EDA_API_KEY}" }
    }
  }
}

Add the server to .vscode/mcp.json. VS Code prompts for the key the first time and stores it for you.

{
  "inputs": [
    {
      "id": "eda-key",
      "type": "promptString",
      "description": "EasyDeploy API key",
      "password": true
    }
  ],
  "servers": {
    "easydeploy-ai": {
      "type": "http",
      "url": "https://mcp.easydeploy.ai/mcp",
      "headers": { "Authorization": "Bearer ${input:eda-key}" }
    }
  }
}

Put this in .vscode/mcp.json. A workspace .mcp.json drops the headers without saying so.

Open Settings, then Connectors, then Add custom connector. Paste the server URL, choose API key authentication and paste your EasyDeploy key.

Browser sign-in from ChatGPT is not available yet. Use an API key.

Point any client or SDK at the same URL over streamable HTTP and send your key as a bearer token. To check a key without a client, ask the server for its tool list.

curl -sS https://mcp.easydeploy.ai/mcp \
  -H "Authorization: Bearer eda_live_your_key_here" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

A client that only speaks the older stdio transport can bridge with npx mcp-remote and the same header.