Add multi-agent orchestration recipe

[GerardoRdz96]

What

A recipe demonstrating multi-agent orchestration using Llama models, featuring:

• Context Layers: 4-layer system (identity → state → relevant → archive) for persistent agent behavior
• Agent Cards: JSON capability descriptors for agent discovery and routing
• Safety Guards: Output validation, cost tracking, and rollback for production reliability
• Phase-Gated Pipeline: Research → Build → Review with quality gates

Why

Llama 4 is positioned for powering agentic systems, but the cookbook doesn't have examples of multi-agent orchestration with context engineering. This recipe fills that gap with production-tested patterns from PA·co, a 22-agent system running in production.

How to Test

pip install -r recipes/multi_agent_orchestration/requirements.txt
export LLAMA_BASE_URL=http://localhost:11434/v1
export LLAMA_MODEL=llama3.2
python recipes/multi_agent_orchestration/orchestrator.py "Build a CLI tool that converts CSV to JSON"

Works with Ollama, vLLM, or any OpenAI-compatible endpoint. Zero heavy dependencies (only requests).

Files

File	What
context_layers.py	4-layer context system with pre-built agent identities
agent_cards.py	Capability discovery and routing (inspired by A2A protocol)
safety_guard.py	Output validation, cost tracking, rollback manager
orchestrator.py	3-agent pipeline with phase gates
requirements.txt	Single dependency: requests

Built by PA·co — A Penguin Alley System.

[meta-cla]

Hi @GerardoRdz96!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks!

[meta-cla]

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!