Use Claude Code with DataDoe MCP

Use Claude Code when you want a local coding session in your repository, with DataDoe as a live Amazon data source for your AI agent.

Prerequisites

• Claude Code (opens in a new tab) installed and signed in.

There are two ways to connect. Pick one:

Option A: Use your Claude connector (no API key)

If you have already added DataDoe as a connector in Claude (Using Claude), you are done. Claude Code automatically loads the connectors from your Claude account when you are signed in with the same account. Run /mcp to confirm that datadoe is listed, then skip to the test step below. If it is not listed, connect with the OAuth command below or use Option B.

No connector yet? You can also sign in with OAuth directly from Claude Code:

1claude mcp add datadoe "https://mcp.datadoe.com/mcp/v1" --transport http

Then run /mcp inside Claude Code, select datadoe, and choose Authenticate. Your browser opens so you can sign in to DataDoe and approve access.

Option B: Use a DataDoe MCP key

Use this when you want a connection that is independent of your Claude account, for example shared team projects, CI, or scripted sessions. Create a key in DataDoe MCP Integrations (opens in a new tab), then run this command in your project root:

1claude mcp add datadoe "https://mcp.datadoe.com/mcp/v1" --transport http --scope project --header "datadoe-mcp-key: YOUR_DATADOE_MCP_KEY"

The server name and URL must come before the --header option. --header accepts multiple values, so anything placed after it is read as another header and the command fails with missing required argument 'name'.

Replace YOUR_DATADOE_MCP_KEY with your DataDoe MCP key.

Choose a scope:

• --scope project (used above) saves the config to .mcp.json in your project, which is useful when you want the team to share the same MCP setup. Do not commit a real key. Reference an environment variable instead (.mcp.json expands ${VAR} syntax, so use "datadoe-mcp-key: ${DATADOE_MCP_KEY}" and export DATADOE_MCP_KEY in your shell), or use Option A.
• --scope user saves it globally for your user, which is useful when you want DataDoe available across all projects.

Reload Claude Code

In an existing Claude Code session, run /mcp to reload the server list. For a new session, start Claude Code as usual. It picks up the config automatically.

Test the connection

Run a prompt that uses DataDoe data, for example:

Show me my Amazon Seller Central sales for the last 7 days.

If Claude Code returns real data, you are connected.