Plotkeep — a chart press that never phones home
This is a submission for the Hacktoberfest Open-Source AI Challenge: Week 2 (Side Hustle). Live demo: https://plotkeep-pi.vercel.app/studio Code: https://github.com/sherotree/plotkeep Prize category: Best Use of Gemma . What I built Plotkeep is a static web app that turns a pasted table into a chart without sending the table anywhere. You paste CSV, TSV, or cells copied from a spreadsheet — or a…
Plotkeep is a static web application that converts a pasted table into a chart without sending the data anywhere. Users can paste CSV, TSV, cell data copied from a spreadsheet or simply a sentence describing the table shape. The app parses the input sheet, asks an open-weight Gemma model running locally for a chart specification (type, encodings, title, and insight), and validates the model's JSON output using Zod.
If the model's output is incorrect, an error is sent back to the model up to three times before the app proceeds with a deterministic heuristic to render the chart using Apache ECharts. The app renders the chart as PNG or SVG and allows follow-up prompts to revise the spec. Plotkeep does not have an account, API, or telemetry. The only network used is the first download of Gemma weights or a call to a local Ollama instance if already installed.
The app is intended for freelancers who already own a GPU-equipped laptop and frequently paste client-related data into chart generators. It charges a one-time $29 fee per operator, allowing multiple machines. The license idea aims to charge for a physical "press" instead of inference. The project's source code is available on GitHub, and a live demo is accessible at plotkeep-pi.vercel.app/studio.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.