better agentic engineering.
Analyze your AI coding assistant usage — any harness, one dashboard.
585890575-d60203ce-c678-4fdd-918e-f6c04bff04df.mp4
AI Engineer Coach reads your local AI session logs and turns them into actionable insights — no data leaves your machine.
- Track progress -- practice scores, weekly trends, daily activity charts
- Detect anti-patterns -- 45 rules across prompt quality, session hygiene, code review, tool mastery, and context management
- Measure output -- AI-generated code volume by language, workspace, model, and harness
- Discover skills -- find repeated prompts and turn them into reusable skills
- Score context health — agentic readiness checks, instruction-file audits, workspace context maps
Screenshots
The extension is not published to a marketplace or Releases page, so you build the .vsix yourself and install it. Pick whichever build path fits your setup.
Prerequisites:
- VS Code
- Dev Containers extension
- Docker or Podman
Steps:
- Clone the repo and open it in VS Code.
- Reopen in container.
- Run:
npm ci
npm run package- Install the generated
.vsix(see Install the built VSIX below).
Prerequisites:
- VS Code
- Node.js and npm
Steps:
git clone https://github.com/microsoft/ai-engineering-coach.git
cd ai-engineering-coach
npm ci
npm run packageThen install the generated .vsix (see below).
macOS / Linux
code --install-extension ai-engineer-coach-*.vsixWindows / PowerShell
code --install-extension (Get-ChildItem . -Filter 'ai-engineer-coach-*.vsix' | Select-Object -First 1).FullNameIf the CLI does not work, install it from the VS Code UI: press Ctrl+Shift+P, type Install from VSIX, then browse to the .vsix file and select it.
After install:
- Open the command palette (
Cmd+Shift+P/Ctrl+Shift+P) - Run AI Engineer Coach: Open Dashboard
- Navigate pages from the sidebar, filter by workspace or harness
The same dashboard also runs as a canvas inside the GitHub Copilot app, so you do not need VS Code to use it.
A canvas is an interactive side panel in the GitHub Copilot app. Rather than replying only in chat, the agent can open a canvas to show rich, task-specific UI that you view and interact with directly while you keep working. Extensions register their own canvases, and this repo ships one named AI Engineer Coach under .github/extensions/ai-engineer-coach/. It reuses the exact webview bundle from the VS Code extension and parses your local session logs in process, so nothing leaves your machine.
To open it:
- Clone this repo and open it as a project in the GitHub Copilot app.
- Build the project once:
npm install && npm run build- Open the AI Engineer Coach canvas. On a fresh clone it shows a setup card with the build command and reloads into the full dashboard once the build finishes. No manual reopen needed.
A few features depend on the local VS Code language model and are hidden in canvas mode: Skill Finder, Learning Center, the Level Up section, and the Context Health AI review. Everything driven by your on-disk logs (Dashboard, Timeline, Coding Moments, Output, Patterns, Anti-Patterns) works the same. App sessions show up as GitHub Copilot App and terminal sessions as GitHub Copilot CLI in the harness breakdown.
| Page | Description |
|---|---|
| Dashboard | Practice scores with week-over-week trends, daily activity chart, top workspace stats |
| Timeline | Gantt-style session timeline with per-day drill-down and overlap detection |
| Coding Moments | Screenshot gallery from AI coding sessions with story reels and workspace filtering |
| Page | Description |
|---|---|
| Output | Generated code volume by language, model usage table (token breakdown temporarily hidden) |
| Burndown | Monthly AI token budget progress with projections (temporarily disabled) |
| Patterns | 7×24 activity heatmap and work-life balance signals |
| Page | Description |
|---|---|
| Anti-Patterns | Five practice score cards with severity ratings, concrete actions, and example prompts. 45 editable markdown rules plus a coverage heatmap |
| Rule Editor | Create, edit, and tune detection rules visually or as raw markdown. Live-test against your data |
| Rule Playground | Interactive REPL for the rule DSL with field browser, function catalog, and metric list |
| Data Explorer | Browse session fields, view distributions, run ad-hoc filters |
| Skill Finder | Discover repeated prompt patterns and matching community skills from the open-source catalog |
| Context Health | Overall context score, agentic readiness checklist, workspace context map, AI-powered instruction-file review |
| Page | Description |
|---|---|
| Learning Center | Personalized quizzes and code-comparison rounds generated from your actual usage |
| Achievements | XP-based progression with Bronze → Silver → Gold → Diamond tiers |
| Agentic SDLC | How you use AI across the full software-development lifecycle |
| Share | Generate a shareable stat card and export Markdown/JSON summaries |
- Read-only — the extension never modifies your session files
- Local analysis — all parsing and analytics run entirely on your machine
- No proprietary telemetry — the extension does not phone home or collect usage data
- Optional AI features — some features (rule compiler, skill finder, context review) use the VS Code built-in Copilot language model API when explicitly invoked by the user
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.
This project is an open-source community effort by Microsoft employees. It is not an official Microsoft product and is not part of any Microsoft service or support offering. It is provided as-is with no warranties or guarantees.