What runtime dependencies does it require?
It requires Python and `uv` package manager. The `mcp_server_datavis` module is invoked via `uv run`.
Data Visualization MCP Server
@isaacwasserman
基本信息
关于 Data Visualization MCP Server
暂无概览
Data Visualization MCP Server 配置
Data Visualization MCP Server 工具
2Save a table of data agregations to the server for later visualization
Visualize a table of data using Vega-Lite syntax
概览
What is Data Visualization MCP Server?
A Model Context Protocol (MCP) server that gives large language models an interface to visualize data using Vega‑Lite syntax. It is designed for use with MCP clients such as Claude Desktop to generate charts and graphs from structured data.
How to use Data Visualization MCP Server?
Add the server to your Claude Desktop configuration (claude_desktop_config.json) by specifying the command uv, the directory path, the Python module mcp_server_datavis, and the --output_type flag (set to either png or text). Use the save_data tool to store a data table and the visualize_data tool to render that data as a Vega‑Lite visualization.
Key features of Data Visualization MCP Server
- Two core tools:
save_dataandvisualize_data - Supports Vega‑Lite specification for chart generation
- Output can be text (Vega‑Lite spec) or PNG image
- Data tables are saved on the server for reuse
- Integrates with MCP clients like Claude Desktop
Use cases of Data Visualization MCP Server
- Generate a PNG chart from aggregated data returned by an LLM
- Iteratively refine a Vega‑Lite specification before rendering
- Create reusable data tables for multiple visualizations in a session
- Embed a Vega‑Lite specification in a text response for rendering elsewhere
常见问题
How are data tables persisted?
Data tables are saved in the server’s memory using the `save_data` tool and are referenced by name in `visualize_data`. No external database is used.
What output formats are supported?
Two output types are supported: `text` (returns a JSON Vega‑Lite specification) and `png` (returns a base64‑encoded PNG image wrapped in an MCP `ImageContent` container).
Is authentication required?
No authentication or authorization mechanism is described in the README. The server runs locally via stdio transport.
What transport protocol does it use?
It uses standard MCP stdio transport, invoked from the command line.
数据与分析 分类下的更多 MCP 服务器
Ultimate Web Scraper
Ultimate Web ScraperExtract products, prices, images and reviews from any page in one click — or connect your AI agent and let it do the scraping for you. Works great for: - Ecommerce: extract every product from a Shopify, WooCommerce or
NeuralVerge
NeuralVerge is the data layer for agentic AI. Ask a question in plain language and get a structured, cited answer drawn from 300+ sources — e-commerce marketplaces, corporate registries, company intelligence, enrichme
Convertlyft
ConvertlyftWebsite analytics and SEO in one. Every visit counted, no sampling. Each to-do says what leaked and the change to make, with the numbers behind it, you can visit us on Convertlyft.com first 5,000 Visits to your website f
Companero – Romanian Company Data
CompaneroRemote MCP server for Romanian company data: search 2.7M Romanian companies by CUI, county, CAEN industry, revenue and founding date; get company profiles, balance sheets, ANAF tax debts, legal representatives and due di
Subtext
Subtext by FullstorySession replay, built for agents. Subtext is agentic session review: it captures production sessions of your app and connects them to your coding agent — Claude Code, Cursor, Codex, Devin, your own harness — so it can
9
评论