AI Class by Kanary

Knight or Ninja? Your Codex & Claude logs decide

OpenAI DayDeveloper ToolsArtificial IntelligenceProductivity

▲ 62 votes2 commentsLaunched Sep 18, 2026

Daily #84Weekly #135Monthly #6740

Not a quiz. AI Class reads 30 days of your real Codex and Claude Code logs and turns them into one of 16 RPG classes: Knight, Ninja, Alchemist, Guild Master and 12 more. Six stats show how you actually work: volume, autonomy, chat length, context size, cache reuse and parallelism. Paste one prompt into your agent. It aggregates your logs locally and sends only totals. Conversations, prompts and project names never leave your machine. Free. No install. Your card in about a minute.

AI Analysis

📝 Summary

AI Class by Kanary analyzes 30 days of Codex and Claude code logs to assign one of 16 RPG classes (e.g. Knight, Ninja, Alchemist) with six stats: volume, autonomy, chat length, context size, cache reuse, parallelism. Core features include local log aggregation (only totals sent), complete privacy (no conversations or prompts leave device), free access, no install, and results in ~1 minute via a single prompt. It solves developers' lack of visibility and fun insights into their AI coding habits. USP is gamified self-reflection on real usage patterns. Value proposition: engaging, private way to understand and optimize AI-assisted workflows.

📈 Market Timing

Favorable in 2025-2026 with explosive growth and mainstream adoption of AI coding tools like Claude and Copilot. Developers increasingly demand insights into usage patterns for productivity gains amid maturing AI tech and focus on responsible AI. Economic push for efficiency and fun personalization tools align well. Excellent Timing.

✅ Feasibility

High. Low technical difficulty via prompt-based local processing with no install or heavy backend. Minimal dev/operation costs as it's free with no user data storage. Built-in privacy reduces compliance risks. Strong scalability for broad agent use. Potential log format challenges exist but overall low risk and high team fit for AI tool builders. High.

🎯 Target Market

Main segments: Software developers, full-stack engineers, indie hackers and AI enthusiasts aged 25-40 heavily using Claude/Codex. Industries: Software development, tech startups. Geographic: Global with concentration in US, Europe, China tech hubs. TAM for AI dev tools ~$20B by 2026; SAM for usage analytics ~$1B; SOM for fun/individual tools ~$100M. Core pains: opaque AI habits and inefficiency. Moderate willingness to pay (currently free, potential freemium).

⚔️ Competition

Low. Direct competitors: 1. GitHub Copilot Usage Insights (github.com/features/copilot), 2. Anthropic Usage Console (console.anthropic.com), 3. Cursor Analytics (cursor.com), 4. LinearB AI Dev Analytics (linearb.io). Advantages: unique RPG gamification, superior privacy via local-only processing, fun UX, free. Disadvantages: lighter on enterprise-grade metrics, may lack depth for power users vs professional analytics platforms.

Upgrade Pro to unlock full AI analysis