| name | copilot-agent-patterns |
| description | GitHub Copilot custom agent design patterns, best practices, collaboration workflows, and agent orchestration |
| license | Apache-2.0 |
Copilot Agent Patterns Skill
🔴 AI FIRST Quality Principle
Apply the AI FIRST principle: never accept first-pass quality. Minimum 2 iterations. Read all output, improve every section. No shortcuts.
Purpose
Provides proven patterns for designing, implementing, and orchestrating GitHub Copilot custom agents.
Agent Design Patterns
Specialist Agent Pattern
- Focused on single domain expertise
- Minimal tool set (
view, edit, create, shell, search_code)
- Clear scope boundaries (MUST/MUST NOT)
- Skills integration references
Orchestrator Agent Pattern
- Broad tool access (
tools: ["*"])
- Creates and assigns GitHub issues
- Delegates to specialist agents
- Tracks progress and quality
YAML Frontmatter Standards
name: agent-name
description: Brief expertise
tools: ["tool1", "tool2"]
Collaboration Patterns
- Sequential: Task agent → creates issues → assigns specialists
- Parallel: Multiple specialists work on different scopes simultaneously
- Stacked PRs: Foundation PR → Implementation PR → Polish PR (using base_ref)
Agent Boundaries
- Each agent respects its expertise area
- Delegates outside its scope
- References skills library constantly
- Follows ISMS policies automatically
Copilot Coding Agent Tools
assign_copilot_to_issue — with base_ref and custom_instructions
create_pull_request_with_copilot — with base_ref and custom_agent
get_copilot_job_status — for tracking progress
Best Practices
- Focus on expertise area
- Complete work autonomously when possible
- Test changes thoroughly
- Document decisions
- Only ask when genuinely ambiguous
Related Policies
🔗 Integration with agentic workflows & analysis artifacts
This skill is consumed by the 11 agentic news workflows in .github/workflows/news-*.md. The authoritative contract lives in .github/prompts/README.md; this skill supplies domain expertise on top of that contract.