az-scout-plugin-strategy-advisor 2026.2.1

Strategy Advisor plugin for az-scout — capacity deployment strategy recommendations

az-scout-plugin-strategy-advisor

⚠️ Work in Progress — This plugin is under active development. APIs, models, and behaviors may change without notice. It is not recommended for production use at this time.

Strategy Advisor plugin for az-scout — capacity deployment strategy recommendations.

Features

• UI tab — Strategy Advisor form with subscription picker, workload constraints, and results rendering
• API route — POST /plugins/strategy/capacity-strategy to compute deployment strategies
• MCP tool — capacity_strategy tool for AI agents to obtain deployment recommendations
• Strategy engine — Evaluates (region, SKU) combinations against zones, quotas, restrictions, spot scores, pricing, confidence, and inter-region latency

Strategies

The engine recommends one of: single_region, active_active, active_passive, sharded_multi_region, burst_overflow, time_window_deploy, progressive_ramp.

Setup

# Install alongside az-scout
uv pip install -e .

# The plugin is auto-discovered at startup
az-scout

Structure

src/az_scout_strategy/
├── __init__.py          # Plugin class + module-level `plugin` instance
├── engine.py            # Capacity strategy computation engine
├── models.py            # Pydantic models (request/response)
├── routes.py            # FastAPI APIRouter (POST /capacity-strategy)
├── tools.py             # MCP tool function
└── static/
    ├── html/
    │   └── strategy-tab.html  # HTML fragment (fetched by JS at runtime)
    └── js/
        └── strategy-tab.js    # Tab UI logic

Dependencies

This plugin depends on the following az-scout core modules:

• az_scout.azure_api — ARM API calls (SKUs, quotas, spot scores, prices)
• az_scout.scoring.deployment_confidence — Confidence scoring
• az_scout.services._evaluation_helpers — Shared evaluation utilities
• az_scout.services.region_latency — Inter-region latency data

Quality checks

uv run ruff check src/ tests/
uv run ruff format --check src/ tests/
uv run mypy src/
uv run pytest

Versioning

Uses CalVer (YYYY.MM.MICRO) via hatch-vcs. Tag and push to release:

git tag v2026.2.0
git push origin v2026.2.0

License

MIT

Disclaimer

This tool is not affiliated with Microsoft. All capacity, pricing, and latency information are indicative and not a guarantee of deployment success. Spot placement scores are probabilistic. Quota values and pricing are dynamic and may change between planning and actual deployment. Latency values are based on Microsoft published statistics and must be validated with in-tenant measurements.