Analysis updated 2026-07-03 · repo last pushed 2026-07-03
A solo developer tracks whether they are over-relying on AI for certain programming languages.
A team lead shares stat cards to compare different AI collaboration approaches across the team.
A developer uses XP and quizzes to gamify and level up their AI prompting skills over time.
A user runs context health checks to see if their workspace is set up well for agentic AI workflows.
| microsoft/ai-engineering-coach | crafter-station/petdex | vercel/eve | |
|---|---|---|---|
| Stars | 3,193 | 3,226 | 3,152 |
| Language | TypeScript | TypeScript | TypeScript |
| Last pushed | 2026-07-03 | 2026-07-01 | 2026-07-03 |
| Maintenance | Active | Active | Active |
| Setup difficulty | moderate | moderate | easy |
| Complexity | 2/5 | 2/5 | 3/5 |
| Audience | developer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Not published to a marketplace, so you must clone and build the extension yourself from source using the provided steps.
AI Engineer Coach is a VS Code extension that analyzes how you use AI coding assistants and gives you a dashboard of insights to help you get better at it. It reads the session logs that tools like GitHub Copilot generate on your machine and turns them into scores, trends, and concrete feedback, all without sending your data anywhere. At a high level, it scans your local AI session files and runs them through 45 detection rules covering things like prompt quality, session hygiene, code review habits, and context management. You get visualizations: a timeline of your coding sessions, charts showing how much code the AI generated by language and model, a heatmap of when you work, and anti-pattern cards that flag specific bad habits with suggested fixes. It also includes a skill finder that spots repeated prompts you could turn into reusable snippets, and a context health score that checks whether your workspace is set up well for agentic AI workflows. Some optional features use VS Code's built-in Copilot model, but only when you explicitly trigger them. The tool is built for developers and teams who use AI coding assistants regularly and want to understand their patterns, measure their output, and improve over time. A solo developer might use it to see if they're over-relying on AI for certain languages or working at unhealthy hours. A team lead could share stat cards to compare approaches. The gamification layer, XP, achievement tiers, personalized quizzes generated from your actual usage, suggests it's also aimed at people who want a structured way to level up their AI collaboration skills rather than just guessing whether they're using these tools well. Privacy is a core design choice: the extension is read-only, never modifies your session files, runs all analysis locally, and includes no telemetry. It also works as a canvas inside the GitHub Copilot app, not just VS Code, though some features are unavailable in that mode. The extension isn't published to a marketplace, so you build it yourself from source, a hurdle for non-developers, but straightforward if you follow the provided steps.
A VS Code extension that analyzes your local AI coding assistant logs to score your prompt quality, flag bad habits, and show dashboards, all processed privately on your machine with no data sent anywhere.
Mainly TypeScript. The stack also includes TypeScript, VS Code Extension API, GitHub Copilot.
Active — commit in last 30 days (last push 2026-07-03).
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.