Skip to content

10-quick-reference-cards.md

Latest commit

 

History

History
487 lines (362 loc) · 13.2 KB

10-quick-reference-cards.md

File metadata and controls

487 lines (362 loc) · 13.2 KB

GitHub Copilot Quick Reference Cards

Document 10 | Cheat Sheets

Sri Bolisetty | December 2025

Purpose

This document provides condensed, printable quick reference cards for each persona. Each card is designed to fit on 1-2 pages and covers the essential information for daily use.

Card Index

Card Audience Focus
QRC-01 Developers Keyboard shortcuts, modes, tips
QRC-02 Engineering Leaders Metrics, adoption, reporting
QRC-03 C-Suite ROI, cost, governance
QRC-04 Security/InfoSec Controls, compliance, risk
QRC-05 Administrators Configuration, troubleshooting

10.1 QRC-01: Developer Quick Reference

GITHUB COPILOT DEVELOPER CARD | Sri Bolisetty 2025

The Mental Model

COPILOT IS AUTOCOMPLETE ON STEROIDS — NOT CHATGPT IN YOUR IDE

  • It finishes your sentences, not answers your questions
  • 80% of value comes from inline completions (Tab key)
  • Context from YOUR code drives suggestions

Essential Keyboard Shortcuts

VS Code

Action Windows/Linux Mac
Accept suggestion Tab Tab
Accept word Ctrl+→ Cmd+→
Next suggestion Alt+] Option+]
Previous suggestion Alt+[ Option+[
Show all suggestions Ctrl+Enter Cmd+Enter
Dismiss Esc Esc
Open Chat Ctrl+Shift+I Cmd+Shift+I
Inline Chat Ctrl+I Cmd+I

JetBrains

Action Windows/Linux Mac
Accept suggestion Tab Tab
Next suggestion Alt+] Option+]
Previous suggestion Alt+[ Option+[
Show all suggestions Alt+Enter Option+Enter
Dismiss Esc Esc
Toggle Copilot Ctrl+Alt+Shift+O Cmd+Option+Shift+O

The Five Modes

Mode Speed Scope When to Use
Inline Instant Line/block 80% of time - flow state
Chat Seconds Question Explanations, questions
Edits Seconds Selection Refactoring
Agent Mode Minutes Multi-file Complex features
Coding Agent Hours Repo Async/overnight tasks

Chat Slash Commands

Command Use For
/explain Understand existing code
/fix Debug and fix issues
/tests Generate test cases
/doc Generate documentation
/optimize Improve performance

Chat Agents

Agent Context Example
@workspace Entire repo "Where is auth handled?"
@terminal Shell "How do I run tests?"
@vscode Editor "How do I multi-select?"

Context Engineering Tips

✅ DO:
  • Name variables clearly (userEmailAddress, not str1)
  • Write descriptive comments before functions
  • Open relevant files in tabs
  • Start patterns for Copilot to follow

❌ DON'T:
  • Spend >30 seconds waiting for suggestions
  • Fight the tool — just type manually
  • Accept without reviewing
  • Expect it to understand architecture

The 3-Second Rule

If you're looking at ghost text for more than 3 seconds, you're doing it wrong. Tab, type, or move on.

10.2 QRC-02: Engineering Leader Quick Reference

GITHUB COPILOT ENGINEERING LEADER CARD | Sri Bolisetty 2025

Key Metrics to Track

Metric Target Where to Find
Weekly Active Users >70% Dashboard → Overview
Acceptance Rate >25% Dashboard → Completions
Agent Mode Adoption 20-40% Dashboard → Agent Metrics
Inactive Users <10% Dashboard → Seats

Dashboard Access

GitHub → Enterprise → Insights → Copilot usage

Note: Data has 3-day delay. Monday data visible Thursday.

Adoption Health Indicators

Status Indicators Action
🟢 Healthy >70% active, >25% acceptance Maintain
🟡 Warning 50-70% active, 15-25% acceptance Engage champions
🔴 Critical <50% active, <15% acceptance Intervention needed

Three-Phase Adoption Framework

PHASE 1: SEED (Weeks 1-2)
├── Select 5-10 champions
├── Train on Foundation + Developer Track
├── Weekly check-ins
└── Document wins and blockers

PHASE 2: GROW (Weeks 3-6)
├── Champions train their teams
├── Peer-to-peer learning
├── Weekly "office hours"
└── Scale to 50-100 developers

PHASE 3: SUSTAIN (Ongoing)
├── Monthly metrics review
├── Quarterly training refreshers
├── Champion rotation
└── Continuous improvement

Quick ROI Calculation

ROI = (Hours Saved × Hourly Cost - License Cost) / License Cost × 100

Example (100 developers):
  4 hrs/week × $100/hr × 100 devs × 52 weeks = $2,080,000 value
  $39/mo × 100 devs × 12 months = $46,800 cost
  ROI = ($2,080,000 - $46,800) / $46,800 = 4,345%

Common Objections & Responses

Objection Response
"It writes bad code" Train on correct mental model; all code still reviewed
"I'm faster without it" 1-week A/B test challenge
"Security concerns" SOC 2/ISO 27001 certified; enterprise controls
"It'll take my job" Augments, doesn't replace; removes drudgery

Monthly Report Template

EXECUTIVE SUMMARY
• Active Users: [N] of [M] ([X]%)
• Trend: [↑/↓] [X]% from last month
• Estimated Value: $[X]

KEY METRICS
• Acceptance Rate: [X]% (target: 25%)
• Agent Mode Adoption: [X]%

ACTION ITEMS
• [Item 1]
• [Item 2]

10.3 QRC-03: C-Suite Quick Reference

GITHUB COPILOT EXECUTIVE CARD | Sri Bolisetty 2025

What is GitHub Copilot?

AI-powered code completion that helps developers write code faster. Like autocomplete for programmers — but dramatically smarter.

Pricing (December 2025)

Plan Cost Best For
Business $19/user/mo Teams without GitHub Enterprise
Enterprise $39/user/mo Orgs with GitHub Enterprise Cloud

Note: Enterprise requires GitHub Enterprise Cloud (~$21/user/mo additional)

ROI Summary

BREAK-EVEN: ~24 minutes saved per developer per week
TYPICAL ROI: 500-4,000% (conservative estimates)
PAYBACK PERIOD: <1 month

Investment Decision Framework

Factor Assessment
Cost $19-39/user/mo
Risk Low (SOC 2, ISO 27001 certified)
Return 500%+ ROI typical
Competitive 77,000+ organizations using
Talent Developers expect AI tools

Security & Compliance

Certification Status
SOC 2 Type 2 ✅
ISO 27001 ✅
CSA STAR Level 2 ✅
Code used for training ❌ No (Business/Enterprise)

Governance Checklist

  • Security review completed
  • Acceptable use policy approved
  • Feature controls configured
  • Metrics reporting established
  • Training program deployed

Key Questions to Ask

1. What is our adoption rate? (Target: >70%)
2. What is our ROI? (Calculate quarterly)
3. Are there security concerns? (Review with CISO)
4. How does this affect talent? (Developer satisfaction)

10.4 QRC-04: Security/InfoSec Quick Reference

GITHUB COPILOT SECURITY CARD | Sri Bolisetty 2025

Data Flow Summary

flowchart LR
    DEV["👨‍💻 Developer<br/>IDE"] --> PROXY["🔒 GitHub<br/>Proxy"]
    PROXY --> LLM["🤖 LLM<br/>(OpenAI)"]
    
    PROXY -.- NOTE["In-memory only<br/>No retention<br/>No training"]
    
    style DEV fill:#d1ecf1,stroke:#17a2b8
    style PROXY fill:#d4edda,stroke:#28a745
    style LLM fill:#fff3cd,stroke:#856404
Loading

Key Security Facts (Business/Enterprise)

Question Answer
Is code stored? No (in-memory only)
Is code used for training? No
What about third-party models? Same rules apply
Data retention (IDE)? Prompts not retained
Data retention (engagement)? 2 years

Required Security Controls

Control Setting Why
Public code filter ENABLED Prevent IP issues
Content exclusion CONFIGURED Protect sensitive code
Audit logging ENABLED Forensics/compliance
SSO/SAML REQUIRED Access control

Content Exclusion Patterns

# Recommended exclusions
"*":
  - "**/.env"
  - "**/*.pem"
  - "**/*.key"
  - "**/secrets/**"
  - "**/credentials.*"

⚠️ WARNING: Agent Mode does NOT respect content exclusions

Compliance Mapping

Framework GitHub Certification
SOC 2 Type 2 ✅ Certified
ISO 27001 ✅ Certified
CSA STAR Level 2 ✅ Certified
TISAX ✅ Certified

Risk Register Quick View

Risk Residual Level Controls
Data transmission Low No retention, encryption
Code quality Medium Human review required
IP exposure Low Public code filter
Credential leak Medium Content exclusion, scanning

Security Review Triggers

Mandatory review when:
• New Copilot features released
• Enabling Agent Mode or Coding Agent
• Enabling third-party models
• Enabling Copilot Extensions
• Annual recertification due

Incident Response Quick Steps

1. IMMEDIATE: Disable for affected users
2. PRESERVE: Export audit logs
3. ASSESS: Scope of exposure
4. CONTAIN: Rotate any exposed credentials
5. REPORT: Notify stakeholders
6. REMEDIATE: Update controls

10.5 QRC-05: Administrator Quick Reference

GITHUB COPILOT ADMINISTRATOR CARD | Sri Bolisetty 2025

Admin Console Navigation

Enterprise → Settings → AI Controls → Copilot

Policy Configuration Paths

Setting Path
Feature policies AI Controls → Copilot → Policies
Model policies AI Controls → Copilot → Models
Content exclusion Organization → Settings → Copilot → Content exclusion
Seat management Organization → Settings → Copilot → Seat management
Metrics Enterprise → Insights → Copilot usage

Recommended Day-1 Settings

Setting Value Rationale
Copilot in IDE Enabled Core functionality
Copilot Chat Enabled Core functionality
Public code filter Enabled IP protection
Agent Mode Disabled initially Enable after review
Copilot Extensions Disabled Per-extension approval
Third-party models Disabled Vendor review needed
Preview features Disabled Stability

Network Allowlist

Domain Purpose
github.com Authentication
api.github.com API calls
*.githubcopilot.com Copilot service
copilot-proxy.githubusercontent.com Suggestions
copilot-telemetry.githubusercontent.com Metrics

Troubleshooting Quick Checks

Issue: Copilot not working
  ☐ Check network allowlist
  ☐ Verify user has license assigned
  ☐ Check extension version
  ☐ Verify SSO authentication
  ☐ Check policy settings

Issue: No metrics showing
  ☐ Metrics enabled at enterprise level?
  ☐ Wait 3+ days for data freshness
  ☐ User has telemetry enabled in IDE?
  ☐ Extension version meets minimum?

Seat Management Commands

# View seats via API
curl -H "Authorization: Bearer TOKEN" \
  "https://api.github.com/orgs/ORG/copilot/billing/seats"

# Cleanup Action (GitHub Action available)
# github.com/austenstone/copilot-license-cleanup

Data Freshness

Dashboard data has 3 FULL UTC DAYS delay
Monday's activity → visible Thursday
Plan reports/reviews accordingly

Support Escalation

Issue Type Escalation Path
User can't connect Check network → Check auth → GitHub Support
Metrics missing Wait 5 days → GitHub Support
Policy not applying Check hierarchy → GitHub Support
Billing issue GitHub Support (immediate)
Security incident SOC → GitHub Support (immediate)
Chat/context issues See Document 06 Escalation Ladder

Extended Troubleshooting

For context corruption, M365 Copilot issues, or complex chat problems: See Document 06: Integration Guide, Sections 6.9-6.10

Quick Links

Resource URL
Trust Center copilot.github.trust.page
Documentation docs.github.com/copilot
Status githubstatus.com
Community github.com/orgs/community/discussions

Printing Guide

Each card is designed to be printed on:

  • Single page: Use 80% scaling
  • Double-sided: Front = main content, Back = reference tables
  • Card stock recommended for durability

Version History

Version Date Changes
1.0 December 2025 Initial cards

Related Documents

Document Relevance
01 - Developer Track Full developer training
02 - Engineering Leaders Track Full leader training
03 - C-Suite Track Full executive training
04 - InfoSec Architecture Track Full security training
05 - Administrators Track Full admin training