Claude.ai (web)

1. Go to claude.ai (Pro, Max, Team, or Enterprise).
2. Open Settings → Connectors.
3. Click Add custom connector.
4. Set Name to SellerSheet.
5. In Remote MCP server URL, paste https://sellersheetai.com/mcp.
6. Click Connect and sign in with Google when prompted.
7. Test — start a new chat and ask for your Amazon sales over the last 7 days.

Claude Desktop (app)

1. Open Settings → Developer → Edit Config (claude_desktop_config.json).
2. Add this server:
   Copy

   {
     "mcpServers": {
       "sellersheet": {
         "type": "http",
         "url": "https://sellersheetai.com/mcp"
       }
     }
   }

3. Fully quit (Cmd+Q / Alt+F4) and reopen Claude Desktop.
4. The first tool call opens a browser for Google sign-in.
5. Test — ask which Amazon stores are connected.

{
  "mcpServers": {
    "sellersheet": {
      "type": "http",
      "url": "https://sellersheetai.com/mcp",
      "headers": { "Authorization": "Bearer <YOUR_API_KEY>" }
    }
  }
}

Cursor

1. Settings → MCP → Add new global MCP server (or edit ~/.cursor/mcp.json; project-level: .cursor/mcp.json):
   Copy

   {
     "mcpServers": {
       "sellersheet": {
         "url": "https://sellersheetai.com/mcp",
         "headers": { "Authorization": "Bearer <YOUR_API_KEY>" }
       }
     }
   }

2. Reload Cursor.
3. Test — ask Cursor to pull your Amazon sales for the last 7 days.

{
  "mcpServers": {
    "sellersheet": {
      "url": "https://sellersheetai.com/mcp"
    }
  }
}

Windsurf

1. Edit ~/.codeium/windsurf/mcp_config.json (Windsurf uses serverUrl, not url):
   Copy

   {
     "mcpServers": {
       "sellersheet": {
         "serverUrl": "https://sellersheetai.com/mcp",
         "headers": { "Authorization": "Bearer <YOUR_API_KEY>" }
       }
     }
   }

2. Restart Windsurf.
3. Test — ask for your current FBA inventory levels.

{
  "mcpServers": {
    "sellersheet": {
      "serverUrl": "https://sellersheetai.com/mcp"
    }
  }
}

Google Antigravity

1. Open … menu → MCP Servers → View raw config (~/.gemini/antigravity/mcp_config.json). Uses serverUrl, like Windsurf:
   Copy

   {
     "mcpServers": {
       "sellersheet": {
         "serverUrl": "https://sellersheetai.com/mcp",
         "headers": { "Authorization": "Bearer <YOUR_API_KEY>" }
       }
     }
   }

2. Reload Antigravity.
3. Test — ask for your Amazon orders from the last 7 days.

{
  "mcpServers": {
    "sellersheet": {
      "serverUrl": "https://sellersheetai.com/mcp"
    }
  }
}

Gemini CLI

1. Run one command (writes to ~/.gemini/settings.json):
   Copy

   gemini mcp add --transport http --scope user \
     -H "Authorization: Bearer <YOUR_API_KEY>" sellersheet https://sellersheetai.com/mcp

2. Run /mcp in Gemini CLI to confirm the server and its tools.
3. Test — ask it to summarize this week's Amazon sales.

gemini mcp add --transport http --scope user sellersheet https://sellersheetai.com/mcp

VS Code (Copilot Chat)

1. Run MCP: Add Server from the Command Palette (or create .vscode/mcp.json):
   Copy

   {
     "servers": {
       "sellersheet": {
         "type": "http",
         "url": "https://sellersheetai.com/mcp",
         "headers": { "Authorization": "Bearer <YOUR_API_KEY>" }
       }
     }
   }

2. Open Copilot Chat in Agent Mode.
3. Test — in Agent Mode, ask for your connected Amazon stores.

{
  "servers": {
    "sellersheet": {
      "type": "http",
      "url": "https://sellersheetai.com/mcp"
    }
  }
}

OpenAI Codex CLI

1. Add to ~/.codex/config.toml (project: .codex/config.toml):
   Copy

   [mcp_servers.sellersheet]
   url = "https://sellersheetai.com/mcp"
   http_headers = { "Authorization" = "Bearer <YOUR_API_KEY>" }
   startup_timeout_sec = 30

2. Restart Codex.
3. Test — ask Codex to list your Amazon stores and SKUs.

Cline (VS Code extension)

1. MCP Servers → Configure MCP Servers → cline_mcp_settings.json:
   Copy

   {
     "mcpServers": {
       "sellersheet": {
         "type": "streamableHttp",
         "url": "https://sellersheetai.com/mcp",
         "headers": { "Authorization": "Bearer <YOUR_API_KEY>" }
       }
     }
   }

2. Click Restart Server.
3. Test — ask Cline to show your recent Amazon orders.

{
  "mcpServers": {
    "sellersheet": {
      "type": "streamableHttp",
      "url": "https://sellersheetai.com/mcp"
    }
  }
}

ChatGPT (Connectors)

1. Use a plan with Apps & connectors — Pro, Plus, Business, or Enterprise.
2. Enable Settings → Apps & Connectors → Advanced → Developer mode.
3. Create a connector: Name SellerSheet, MCP Server URL https://sellersheetai.com/mcp, Auth Custom Header → Authorization: Bearer <YOUR_API_KEY>.
4. Add SellerSheet to a chat (Deep Research / Agent), then test — ask for your Amazon sales this week.

Coze (ByteDance)

1. In your bot, open Skills → MCP Tool → Add.
2. Choose Connect via URL and enter https://sellersheetai.com/mcp.
3. Add a header Authorization: Bearer <YOUR_API_KEY> (Coze is cloud-hosted, so it uses a key, not OAuth).
4. Save and publish the bot, then test — ask it to list your Amazon stores.

Same key-header pattern works for n8n, Dify, Make, and Zapier.

OpenClaw

1. Export your key (OpenClaw is headless, so it uses a key, not OAuth):
   Copy

   export SELLERSHEET_MCP_URL="https://sellersheetai.com/mcp"
   export SELLERSHEET_API_KEY="<YOUR_API_KEY>"

2. Restart the OpenClaw gateway — it reads these on start. (Or register the server with an Authorization: Bearer $SELLERSHEET_API_KEY header.)
3. Test — ask it to list your Amazon stores.

Any Other MCP Client

Any tool speaking MCP streamable HTTP works. Use these settings:

Test a key with curl:

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