GitHub Copilot Insights

Understand and reduce your GitHub Copilot AI‑credit usage — right inside VS Code.

Overview

GitHub Copilot Insights gives you a fast, local view of your Copilot plan, AI credits, and usage habits — and helps you spend fewer credits without changing what you build.

It answers practical questions such as:

• How many AI credits do I have left, and when do they reset?
• Am I burning through credits faster than usual?
• Which models and projects are costing me the most?
• Which of my prompts and sessions are the expensive ones?
• Where am I wasting credits — and what should I do differently?

Everything runs on your machine. No usage data leaves your computer.

What you get

The extension adds a GitHub Copilot Insights view to the activity bar, plus a one‑click Full View that opens the whole dashboard in an editor tab. Six tabs:

Overview

Your AI‑credit quota at a glance — remaining, used, total, percentage, a colored progress bar, and a friendly "mood" indicator. Plus pacing guidance (how much you can spend per day to last until reset), reset countdown and date, a daily‑credits‑used chart for the current month, daily‑capacity estimates by model cost, plan details, and your access details. Copy a summary or the raw data to the clipboard.

Promptimizer · Beta

A zero‑cost "pre‑flight" pad for your prompts. Draft a prompt and it analyzes it before you send — nothing is transmitted, no credits spent. You get an estimated cost for lightweight vs. powerful models, a recommended model, a one‑shot quality checklist (context / expected output / constraints / example), and concrete findings with fixes. Start from a good template, auto‑fix obvious credit drains, and send the prompt straight to Copilot Chat when you're ready.

Model & Project

See where your credits go: usage by model (with a side‑by‑side model comparison), usage by project, a chat‑mode split (agent / ask / edit), and the tools your agent sessions use.

Sessions & Prompts

Browse your Copilot Chat and CLI sessions and your most‑expensive prompts, with search, sort, pagination, and click‑to‑expand drill‑down into each session's turns.

Insights

A practice report: an overall efficiency score, Burn Rate Analysis (recent vs. average burn rate, projected monthly cost, and trend), practice‑area scorecards that show your slips and the fix, a context‑health checklist, notable items for the period, and a rule‑coverage summary.

Best Practices

A handy reference for spending fewer credits — the do's and don'ts, the two model tiers and when to use each, developer and QA playbooks, and the most common credit drains to avoid.

Highlights

• A status‑bar indicator showing your remaining AI credits, with several visual styles.
• Date filters (Current Month / 1d / 7d / 30d) across the data views.
• Local snapshot history that powers burn‑rate and trend analysis over time.
• One‑click export of a summary (Markdown) or the raw data (JSON).
• A sidebar view and a full‑width editor‑tab dashboard — your choice.

Privacy

• Local only. All parsing and analysis run on your machine.
• The extension reads your Copilot plan/quota and your local Copilot Chat & CLI history to build these views. Nothing is sent to any external service by this extension.

Getting started

1. Install the extension.
2. Open the GitHub Copilot Insights icon in the activity bar.
3. Sign in with GitHub if VS Code prompts for authentication.
4. Review your credits, usage, and insights — and use Refresh any time for the latest.

Settings

Search for "GitHub Copilot Insights" in VS Code Settings (or use the gear in the view title bar) to configure the status‑bar style and placement, polling interval, progress‑bar mode, mood indicator, and an optional custom credit limit.

Unofficial, third‑party extension. Not affiliated with or endorsed by GitHub or Microsoft.