Open-Source 'evident-charts' Skill Teaches AI Coding Agents to Make Honest Charts
randal_olson · x · 2026-09-29
Data scientist Randy Olson open-sourced evident-charts (49 GitHub stars), a chart skill for AI coding agents: it guides Claude Code, Codex, Cursor and others to make clear, honest charts, and critiques each one before you see it.
- Usage: just ask for a chart as usual — the skill auto-loads whenever the agent creates or reviews one; you can also attach a PNG/script and ask it to "critique this chart and fix it."
- Checks: data issues like totals mixed with parts, duplicate rows, placeholder codes, preliminary months, plus misleading choices like raw counts on log scales and meaningless rainbow colors — it critiques then rebuilds.
- Olson tested it on Artificial Analysis model-eval data, arguing "one question per chart" beats AA's all-labs-at-once chart — e.g., which lab has held the top Intelligence Index? Anthropic, every day since April 2026.
More from coding & agent
- Garry Tan Backs AI Browsers Like AsideAI: Agent Password Management Is the Key Layer — garrytan · 2026-09-29
- Meeting Prep Agent Separates App State in SQLite from Long-Term Memory — Datrika_Nandhini · 2026-09-29
- Chestnut unveils 18-DOF Aero Hand plus matching exoskeleton for zero-gap humanoid data capture — chris_j_paxton · 2026-09-29
- RubyLLM hits 1.1M downloads in a month, 13M total — kieranklaassen · 2026-09-29
- Incident Memory Agent Turns Postmortems into Verified Triage Context via Hindsight — nagakeerthan · 2026-09-29
- InstaCloud launches agent-native serverless cloud that lets Claude Code and Cursor provision infra via one command — testingcatalog · 2026-09-29