Chart Canvas
Let your coding agent build charts and run queries on SQLite, CSV, Parquet, or JSON without writing a one-off dashboard script.
Overview
Chart Canvas is an MCP server for the Grow phase that creates interactive visualizations and queries SQLite, CSV, Parquet, and JSON data sources from your agent.
What is this MCP server?
- Creates interactive visualizations from agent-driven specs
- Queries SQLite, CSV, Parquet, and JSON data sources
- Ships as @gluip/chart-canvas-mcp via npm stdio transport (v0.3.2)
- Backend repo under gluip/chart-canvas for self-hosted or local datasets
- Package version 0.3.2 (@gluip/chart-canvas-mcp, stdio)
- Supports 4 source types: SQLite, CSV, Parquet, JSON
Community signal: 1 GitHub stars.
What problem does it solve?
You have metrics in files or SQLite but no fast way for your agent to chart and query them without custom Python or a separate BI login.
Who is it for?
Indie SaaS builders with SQLite or CSV/Parquet exports who want agent-driven charts during weekly metrics reviews.
Skip if: Teams that need governed warehouse SQL, row-level security, or a full hosted BI platform instead of local MCP queries.
What do I get? / Deliverables
Your agent can produce interactive charts and run queries on local or project data sources directly through MCP tools.
- Interactive chart outputs from agent requests
- Query results against configured file or SQLite sources
- Repeatable agent workflow for local analytics without custom scripts
Recommended MCP Servers
Journey fit
Solo builders need quick insight from product and ops data after something is running; analytics is the natural shelf for visualization and ad hoc querying. Interactive charts and file/database queries are core analytics workflows for understanding metrics, funnels, and exports.
How it compares
MCP data-viz and query bridge, not a hosted analytics dashboard or a static charting skill.
Common Questions / FAQ
Who is Chart Canvas for?
Solo and indie builders who keep product data in SQLite, CSV, Parquet, or JSON and want their AI coding agent to visualize and query it without a separate analytics app.
When should I use Chart Canvas?
Use it in the Grow phase when you need quick charts, exploratory queries, or demo visuals from datasets you already have on disk or in a dev database.
How do I add Chart Canvas to my agent?
Install @gluip/chart-canvas-mcp from npm, add it as a stdio MCP server in Claude Code or Cursor, and point tool calls at your SQLite path or file URLs your environment can read.