FAOS optimize preparation skill

[Luffy Chen (XOEEst)]

This pull request introduces a new FAOS optimization workflow for Python agents in the Microsoft Foundry skill set. It adds comprehensive documentation and usage patterns for converting existing Python agent code to be optimization-ready, including a step-by-step guide, best practices, and framework-neutral code patterns. The documentation ensures that users can expose agent configuration knobs for optimization, review changes before deployment, and preserve existing runtime behavior.

Key additions and changes:

New FAOS Optimization Workflow Documentation

• Added a new sub-skill, faos-optimize, to the Foundry agent skills, enabling users to convert Python agent code for FAOS optimization with a required review step before deployment. [1] [2]
• Created detailed documentation in faos-optimize.md, outlining a 12-step workflow for identifying, wiring, and exposing configuration knobs (like instructions, model, and temperature) for FAOS, with guardrails and review checkpoints to ensure safe and targeted changes.
• Provided a quick reference table summarizing supported languages, required patterns and knobs, and review/deployment workflow.

Python Implementation Patterns

• Added python-patterns.md, detailing framework-neutral code snippets and best practices for wiring FAOS optimization knobs in single-agent and multi-agent Python architectures, including guidance for Microsoft Agent Framework, FastAPI, LangGraph, and custom runtimes.
• Included environment variable and dependency management recommendations, as well as a verification checklist to ensure agent integrity after conversion.

User Guidance and Workflow Integration

• Updated the main skill documentation to include the new faos-optimize workflow and clarified when to use it in the agent lifecycle, integrating it into the user intent-to-workflow mapping. [1] [2]

These changes provide clear, actionable guidance for making Python agents optimization-ready within the Microsoft Foundry ecosystem, ensuring best practices and safe adoption of FAOS optimization features.## Description

Checklist

• Tests pass locally (cd tests && npm test)
• If modifying skill descriptions: verified routing correctness with integration tests (npm run test:skills:integration -- <skill>)
• If modifying skill USE FOR / DO NOT USE FOR / PREFER OVER clauses: confirmed no routing regressions for competing skills

Related Issues

[Copilot]

Pull request overview

Adds a new faos-optimize workflow under the microsoft-foundry skill to guide users through converting existing Python agent code into a FAOS optimization-ready form, with a mandatory review checkpoint before deployment.

Changes:

• Registers a new faos-optimize sub-skill in the main microsoft-foundry skill and intent-to-workflow routing table.
• Adds a 12-step faos-optimize workflow document covering targeting, wiring config knobs, verification, and review gating.
• Adds a framework-neutral Python patterns reference for implementing the FAOS config contract across common Python agent architectures.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
plugin/skills/microsoft-foundry/SKILL.md	Adds faos-optimize to the sub-skill list and agent lifecycle routing table.
plugin/skills/microsoft-foundry/foundry-agent/faos-optimize/faos-optimize.md	New end-to-end FAOS optimization-readiness workflow doc (with review gate before deploy).
plugin/skills/microsoft-foundry/foundry-agent/faos-optimize/references/python-patterns.md	New Python implementation patterns reference for wiring FAOS knobs while preserving runtime behavior.

[Jon Gallant (jongio)]

One item the bot didn't catch: FAOS is never expanded anywhere in the new docs. Worth defining once at the top so the agent can answer if a user asks what it stands for.