Datawrapper

Datawrapper

A Model Context Protocol (MCP) server for creating Datawrapper charts using AI assistants.

10
Stars
5
Forks
15
Releases

Overview

This MCP server integrates with the Datawrapper Python library to let AI assistants manage Datawrapper charts entirely through conversation. It uses Pydantic validation to ensure inputs are well-formed and interacts with the Datawrapper API using a token-based authentication (DATAWRAPPER_ACCESS_TOKEN). Typical flow: create a chart from user-provided data and chart type, publish it to obtain a public URL, update data or styling (for example, changing the line color), retrieve the chart editor URL for manual editing, and embed the resulting PNG image in responses. The server is designed for flexible deployment: it can be run via uvx with a configuration file, installed via pip, or containerized for Kubernetes. The README provides deployment instructions and environment variables, including DATAWRAPPER_ACCESS_TOKEN, MCP_SERVER_HOST, MCP_SERVER_PORT, and MCP_SERVER_NAME. A built-in /healthz endpoint supports Kubernetes health checks, and sample configuration snippets are provided for common deployment methods.

Details

Owner
palewire
Language
Python
License
MIT License
Updated
2025-12-07

Features

Create Datawrapper charts from prompts

Accepts user-provided data and chart type to create a new Datawrapper chart via the MCP server.

Publish charts and obtain public URL

Publishes the created chart and returns its public URL for sharing.

Update chart data

Allows updating the chart data and basic configuration through chat.

Update chart styling

Supports styling changes like line color (e.g., setting to dodger blue).

Editor URL retrieval

Returns the Datawrapper editor URL to view/edit the chart.

PNG embedding

Embeds a PNG image of the chart in responses.

Input validation with Pydantic

Employs Pydantic to validate incoming requests before interacting with Datawrapper.

Token-based authentication and Datawrapper library

Uses Datawrapper API tokens from environment variables and relies on the Datawrapper Python library for API calls.

Audience

AI assistantsEnable AI assistants to create, publish, update, and display Datawrapper charts via chat.
DevelopersIntegrate Datawrapper charting into MCP-enabled automation pipelines and services across teams.

Tags

DatawrapperMCPAI chartingPythonPydanticToken-based authenticationKubernetesDockerChat-driven