microsoft/AI-Engineering-Coach

What it solves

AI Engineer Coach is designed to help developers improve their "agentic engineering" skills by analyzing how they interact with AI coding assistants. It transforms local session logs into a dashboard of actionable insights, allowing users to track their progress, identify bad habits (anti-patterns) in prompting and context management, and measure the actual volume of AI-generated code.

How it works

The tool operates as a VS Code extension or a canvas within the GitHub Copilot app. It reads local AI session logs in a read-only manner, performing all analysis and parsing entirely on the user's machine to ensure privacy. It uses a set of 45 rules to detect anti-patterns across prompt quality and tool mastery, and can optionally use the built-in VS Code Copilot language model API for advanced features like skill discovery and context health reviews.

Who it’s for

Developers who use AI coding assistants and want to quantify their productivity, refine their prompting techniques, and optimize how they provide context to AI agents.

Highlights

• Anti-Pattern Detection: Identifies 45 specific rules across prompt quality, session hygiene, and context management with severity ratings and concrete actions.
• Local-First Privacy: All analytics and parsing run locally; no data leaves the machine.
• Context Health Scoring: Provides agentic readiness checks and workspace context maps to evaluate how well the AI is being guided.
• Skill Finder: Discovers repeated prompt patterns to turn them into reusable skills.
• Comprehensive Metrics: Tracks code volume by language and model, activity heatmaps, and session timelines.
• Gamification: Includes an XP-based progression system with tiers (Bronze to Diamond) and a personalized Learning Center.