Add shared workflow for Microsoft Amplifier CLI

[Copilot]

Adds support for Microsoft Amplifier, an AI-powered modular development assistant with specialized agents for system design, debugging, code exploration, and research.

Implementation

• Shared workflow: .github/workflows/shared/mcp/amplifier.md

  • Three-step setup: UV package manager → Amplifier CLI → AI provider auto-detection
  • Bash tool access: amplifier * and uv *
  • Multi-provider support: Anthropic (recommended), OpenAI, Azure OpenAI, Ollama
  • Auto-configures based on available secrets (ANTHROPIC_API_KEY, OPENAI_API_KEY, AZURE_OPENAI_ENDPOINT)

• Specialized agents: zen-architect, bug-hunter, explorer, web-research, modular-builder, git-ops

• Bundle system: foundation (default), recipes, design-intelligence

• Session persistence: Project-scoped sessions with resume capabilities

Usage

---
name: my-workflow
engine: copilot
imports:
  - shared/mcp/amplifier.md
permissions:
  contents: read
  issues: read
  pull-requests: read
---

Use amplifier's zen-architect agent to design a scalable caching layer.

The agent can invoke amplifier run, amplifier continue, amplifier session list, and manage bundles/agents through the CLI.

References

• GitHub: https://github.com/microsoft/amplifier
• Documentation: https://microsoft.github.io/amplifier-docs/

Original prompt

Add a shared agentic workflow that add support for amplifier (https://github.com/microsoft/amplifier)

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[github-actions]

🔍 PR Triage Results

Category: feature | Risk: high | Priority: 40/100

Scores Breakdown

• Impact: 30/50 - New feature adding Microsoft Amplifier CLI support
• Urgency: 6/30 - Recently created (0.4 hours old), feature enhancement
• Quality: 4/20 - Draft PR, CI pending, multiple commits

📋 Recommended Action: batch_review

This PR adds new functionality for Microsoft Amplifier integration. It should be reviewed in batch with other similar PRs for efficiency. The high risk level warrants careful review of the implementation.

---

Triaged by PR Triage Agent on 2026-02-05

AI generated by PR Triage Agent