Anthropics February 2026 AI Fluency Index studied 9,830 Claude conversations and classified 11 observable collaboration behaviors across all of them. I read the paper and immediately wanted to know where my own sessions landed — not against some abstract ideal, but against that actual population baseline.
Built skill-tree to answer that. It reads your Claude Code or Cowork session history, runs the same 11-behavior classification the Index uses, and shows you which behaviors appear in your sessions versus which you never touch. The three axes it scores are Description, Discernment, and Delegation (the fourth axis, Diligence, isn't recoverable from chat logs).
The part I found most useful: it picks one behavior you haven't tried and surfaces it as a growth quest that persists into your next session via SessionStart hook. Not a summary of what you're already doing — a specific thing you haven't done yet.
Results come back as a tarot-style archetype card with a stable URL (rendered and stored on Fly.io). The classifier runs on Claude Haiku; the full 7-step pipeline — find session files, extract user messages, classify, assign archetype, synthesize narrative, render, return URL — takes 30–60 seconds. Live example at skill-tree-ai.fly.dev/fixture/illuminator.
Install in Claude Code:
claude plugin marketplace add robertnowell/ai-fluency-skill-cards && claude plugin install skill-tree-ai@ai-fluency-skill-cards
Also works in Cowork via skill-tree-ai.zip, and as an MCP server (npm install skill-tree-ai) for Cursor, VS Code, or Windsurf.
Top comments (0)