AI Dev Tools Sync

Synchronize Claude Code and GitHub Copilot configurations to maintain feature parity.

When to Use

Activate this skill when:

• User asks to update Claude Code or Copilot setup
• User wants to add/modify skills, prompts, agents, or instructions
• User wants both tools to work similarly
• User asks about AI dev tool configuration

Quick Reference

.claude/skills/

.github/skills/

.claude/commands/

.github/prompts/

.github/agents/

.claude/workflows/

.github/

.github/instructions/

.github/chatmodes/

GitHub Copilot Features Catalog

Complete catalog of GitHub Copilot customization features (as of 2026).

Configuration Files

Repository-Level (

.github/

)

copilot-instructions.md

instructions/*.instructions.md

applyTo

prompts/*.prompt.md

agents/*.md

skills/*/SKILL.md

chatmodes/*.chatmode.md

AGENTS.md

Also Supported (Backward Compatibility)

• .claude/skills/

  - Copilot auto-reads Claude skills

• CLAUDE.md

  - Copilot reads if present

• GEMINI.md

  - Gemini CLI format

Feature Details

Custom Instructions (

copilot-instructions.md

)

Root instructions auto-included in every request.

# Project Guidelines
- Use TypeScript for all new files
- Follow BEM naming for CSS classes

Path-Scoped Instructions (

.github/instructions/

)

Apply to specific file patterns via

applyTo

---
applyTo: "src/Services/**/*.cs"
excludeAgent: ["code-review"] # Optional: exclude specific agents
---
# Backend C# Guidelines
Use PlatformValidationResult for validation...

Prompts (

.github/prompts/

)

Reusable via

/prompt-name

---
mode: agent # Optional: agent, chat, edit
---
# Debug this issue
1. Analyze the error
2. Find root cause
3. Propose fix

Agent Skills (

.github/skills/

)

Folder structure with SKILL.md + bundled resources:

skills/my-skill/
├── SKILL.md
├── scripts/
├── references/
└── assets/

Chat Modes (

.github/chatmodes/

)

Custom chat personalities with tool restrictions:

---
name: security-reviewer
tools: ["read", "grep", "glob"] # Restrict tools
---
# Security Review Mode
Focus only on security vulnerabilities...

Agents (

.github/agents/

)

Specialized agent definitions:

---
name: frontend-developer
description: UI/UX implementation specialist
---
# Frontend Developer Agent
Specializes in React, TypeScript, CSS...

Sync Patterns

Patterns for keeping Claude Code and GitHub Copilot configurations synchronized.

Skill Sync

Both platforms support skills in

.github/skills/

.claude/skills/

Best Practice: Create skills in

.github/skills/

.github/skills/my-skill/
├── SKILL.md          # Both platforms read
├── scripts/          # Executable scripts
├── references/       # Progressive disclosure docs
└── assets/           # Templates, images

Prompt/Command Sync

.claude/commands/my-cmd.md

.github/prompts/my-cmd.prompt.md

/my-cmd

/my-cmd

#prompt:my-cmd

Sync Strategy:

1. Create in

   .github/prompts/

   (both read)
2. Or maintain duplicates if behavior differs

Instructions Sync

CLAUDE.md

.github/copilot-instructions.md

applyTo

Sync Strategy:

1. Keep core rules in both files
2. Use

   .github/instructions/

   for path-scoped rules (Copilot-specific)
3. Reference detailed docs from both files

Agent Sync

Both platforms read

.github/agents/*.md

Single Source: Maintain agents in

.github/agents/

Workflow Sync

Claude has workflow orchestration. Copilot uses chat modes.

.claude/workflows/

.github/chatmodes/

Decision Matrix

.github/skills/

.github/prompts/

.github/agents/

.claude/workflows/

.github/chatmodes/

Common Sync Tasks

Add New Skill

mkdir -p .github/skills/new-skill
# Create SKILL.md with frontmatter
# Add references/ and scripts/ as needed

Add New Prompt

# Create in .github/prompts/ for both platforms
touch .github/prompts/new-prompt.prompt.md

Update Core Instructions

1. Edit

   CLAUDE.md

2. Edit

   .github/copilot-instructions.md

3. Keep essential rules in sync

Add Path-Scoped Instructions (Copilot)

# Copilot-specific feature
touch .github/instructions/backend-cqrs.instructions.md
# Add applyTo: "src/Services/**/*Command*.cs"

Sync Process

Step 1: Understand Current Setup

Read these files to understand current configuration:

.claude/workflows/orchestration-protocol.md
.claude/workflows/primary-workflow.md
.github/copilot-instructions.md
.github/AGENTS.md
CLAUDE.md

Step 2: Research Latest Features

Search web for:

• "GitHub Copilot features setup 2026"
• "GitHub Copilot custom instructions agents skills prompts"
• "GitHub Copilot agent mode workspace context"

See the GitHub Copilot Features Catalog section above for the full feature catalog.

Step 3: Identify Sync Opportunities

Compare capabilities and identify gaps:

• Skills missing in one platform
• Inconsistent prompt/instruction behavior
• Agent definitions that differ

Step 4: Implement Changes

For each change:

1. Skills: Create in both

   .claude/skills/

   and

   .github/skills/

2. Prompts: Create in both

   .claude/commands/

   and

   .github/prompts/

3. Instructions: Update both

   CLAUDE.md

   and

   .github/copilot-instructions.md

4. Agents: Update

   .github/agents/

   (shared by both)

Compatibility Notes

• Copilot reads

  .claude/skills/

  automatically (backward compatibility)
• Both read

  .github/prompts/*.prompt.md

• Both read

  .github/agents/*.md

• Both read

  AGENTS.md

  in root or

  .github/

• Both support path-based instruction files via

  applyTo

  in frontmatter

Sources

• GitHub Copilot Docs
• Custom Instructions
• Agent Skills
• Awesome Copilot

IMPORTANT Task Planning Notes

• Always plan and break many small todo tasks
• Always add a final review todo task to review the works done at the end to find any fix or enhancement needed