Summary

Proposal to add a comprehensive Software Engineering Team collection with 7 specialized agents for enterprise software development, based on learnings from The AI-Native Engineering Flow experiments at Microsoft.

Motivation

Current awesome-copilot agents cover many individual domains but lack a cohesive collection for end-to-end enterprise software development workflows. This collection fills gaps in:

• UX research and Jobs-to-be-Done analysis
• Technical content creation (vs. content evaluation)
• CI/CD debugging (vs. incident response)
• Product management guidance (vs. PRD generation)
• Comprehensive security including AI/ML threats
• Architecture review (vs. diagram creation)
• Responsible AI practices (broader than accessibility alone)

Proposed Collection

Collection: software-engineering-team.collection.yml

7 Agents (all prefixed with se- for collection identification):

1. SE: UX Designer - Jobs-to-be-Done analysis, user journey mapping, Figma-ready UX research artifacts
2. SE: Tech Writer - Technical documentation, blogs, tutorials, and educational content creation
3. SE: DevOps/CI - CI/CD pipeline debugging, deployment troubleshooting, GitOps workflows
4. SE: Product Advisor - GitHub issue templates, product guidance, requirements refinement
5. SE: Responsible AI - Bias testing, accessibility (WCAG), ethical AI practices, privacy
6. SE: Architecture - Architecture reviews, ADRs, Well-Architected Framework, AI-specific patterns
7. SE: Security - OWASP Top 10 + OWASP LLM Top 10 + OWASP ML Security + Zero Trust

Design Principles

• Standalone: Each agent works independently with no cross-dependencies
• Concise names: Short display names for GitHub Copilot dropdown UX ("SE: UX Designer", etc.)
• Enterprise-ready: Focus on production software development practices
• Complementary: Fills gaps in existing awesome-copilot agents without duplication

Tags

team, enterprise, security, devops, ux, architecture, product, ai-ethics

Related

• Source repository: https://github.com/niksacdev/engineering-team-agents
• Medium article: https://medium.com/data-science-at-microsoft/the-ai-native-engineering-flow-5de5ffd7d877