Our take
Activeregular commits- Activity
- 52
- Community
- 70
- Issues
- 69
- Pull requests
- 48
+62 stars in the last 30 days
- Merges more pull requests than 76% of the projects we track
This is a capable local analytics tool for AI-assisted development, with active development and broad contribution, but users must build and install it themselves.
Good fit if
- You use VS Code or the GitHub Copilot app and want a single view of local coding-assistant sessions.
- You value read-only, on-device analysis and are comfortable building a VS Code extension from source.
- You want editable practice rules and the option to use Copilot-powered features.
Look elsewhere if
- You need a marketplace-published extension or a ready-made release, since the README says installation requires building the VSIX.
- You need all features to work without GitHub Copilot, since some AI-assisted capabilities depend on its language model API.
- You need token-budget tracking now, since the README says the Burndown page is temporarily disabled.
All health signals
| Last commit | 2026-11-03 (3 days ago) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 0 |
| Contributors | 23 (top contributor: 28% of commits) |
| Issues closed, last 90 days | 15+ (typically closed in 2 days) |
| Pull requests merged, last 90 days | 54 (typically merged in 2 days) |
| Project age | 6 months |
Checked on 2026-11-07 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
4,401Forks
617Change since Nov 16, 2026: +1 stars·0 forks
Overview
AI Engineer Coach is a TypeScript-based VS Code extension, with a canvas for the GitHub Copilot app, that analyzes local AI coding assistant session logs. It brings activity from different harnesses into one dashboard so developers can understand their usage, find habits to improve, and track progress without uploading session data.
It is aimed at people who want to assess how they work with AI coding tools, rather than use another coding assistant. Most analysis runs locally; some optional learning and review features use GitHub Copilot's Auto model through the VS Code language model API.
Key Features
- Combines local session activity across supported harnesses in a dashboard, timeline, and daily activity views.
- Detects anti-patterns with editable rules covering prompt quality, session hygiene, code review, tool use, and context management.
- Breaks down AI-generated code volume by language, workspace, model, and harness.
- Surfaces repeated prompts that may be turned into reusable skills.
- Reviews context health with readiness checks, instruction-file audits, and workspace context maps.
- Offers learning quizzes, achievements, and an overview of AI use across the software development lifecycle.
- Includes GitHub Copilot app productivity and issue-credit views when the local app is installed.
Use Cases
- A developer wants to review weekly AI coding habits and identify areas to improve.
- A team member needs to compare AI-assisted output across workspaces, languages, or harnesses.
- Someone maintaining project instructions wants to check workspace context and agentic readiness.
- A GitHub Copilot app user wants to examine local issue and pull-request activity alongside coding sessions.
What you need
Detected in the repository
- Node.js (from package.json)
- A test suite and automated checks on GitHub Actions
License in plain words
MITpermissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the copyright and license notice
- Share your changes: no
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
Requires VS Code, Node.js, and npm for a local build:
git clone https://github.com/microsoft/ai-engineering-coach.git
cd ai-engineering-coach
npm ci
npm run package
Install the generated .vsix in VS Code, then run AI Engineer Coach: Open Dashboard from the command palette. See the README for Dev Container and GitHub Copilot app setup details.
Alternatives
- sniffly: Sniffly analyzes Claude Code logs for usage patterns and message history, while AI-Engineering-Coach focuses on practice patterns across local AI coding sessions.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| AI-Engineering-Coach | TypeScript | MIT | 4.4k | Active |
| sniffly | Python | MIT | 1.3k | Inactive |
Considerations
- The project is actively maintained, with regular recent commits and issue and pull-request activity. It has contributors beyond its leading contributor, so development is not concentrated in one person.
- There are no published releases reported, and the extension is not available through a marketplace or Releases page. Adoption therefore involves building and installing it yourself.
- VS Code 1.115 or later is indicated by the README. The Copilot app canvas requires building the project, and some canvas features are unavailable there.
- Session analysis is read-only and local, but optional rule generation, skill discovery, learning content, context review, and anti-pattern explanations can call GitHub Copilot's Auto model through the VS Code API.
- The Output page's token breakdown is temporarily hidden, and Burndown is temporarily disabled. Issue-credit figures are described as rough estimates, not billing data.
- The repository uses the MIT license. Review its terms and the project's disclaimer before incorporating it into a product or workflow.
Found this useful?
Share it with someone who would like AI-Engineering-Coach.