Skip to main content
GuideServerConfigure and deploy Tabsdata servers on your machine.TutorialsConfigure data integration workflows within a running Tabsdata server.Advanced TutorialsBuild end-to-end workflows between two specific systems.API ReferenceCLI ReferenceRelease Notes
Version: 2.0.0

Connect an AI Agent

Tabsdata's agentic workflow uses an MCP server so a coding agent builds and runs your workflows for you. Install it into your agent before continuing to QuickStart. Claude Code and Codex are currently supported.

Step 1Install the MCP server

Start the agent from a shell whose virtual environment has Tabsdata installed.

Temporary, for 2.0.0

The MCP server needs tabsdata-agent, which ships as its own package and is not yet pulled in by tabsdata[all]. For 2.0.0 it also has to be installed from the Tabsdata package index with the version pinned.

Set the index first. This is needed whichever installer you use:

export PIP_EXTRA_INDEX_URL=https://pypi.tabsdata.com/simple/
export UV_INDEX=https://pypi.tabsdata.com/simple/
export UV_INDEX_STRATEGY=unsafe-best-match

Then install the Tabsdata agent if you have not done so yet:

uv pip install tabsdata-agent==2.0.0

Install Tabsdata covers this as part of the full install, so skip this if you followed that page. These additional steps will not be required once Tabsdata 2.0.0 is on PyPI

claude mcp add tabsdata -- python -m tabsdatak.ag.mcp -
Step 2Using the agent

You need an active session with the Tabsdata server before starting your agent. Log in, then start it from the same terminal:

tdk login --server localhost --user admin --password tabsdata
claude

The agent authenticates through your session, so it can only do what you're authorized to do.

Next: Run Your First Data Integration. Build a working flow end to end.