Educators Guide: Teaching with FCC¶

This guide provides a complete 12-week university course syllabus for teaching multi-agent workflow design, AI governance, and collaborative system engineering using the FCC (Find, Create, Critique) Agent Team Framework.

Course Overview¶

Title: Multi-Agent Workflow Engineering with FCC

Level: Graduate or advanced undergraduate (CS, Software Engineering, or AI)

Prerequisites: Python programming, basic software architecture concepts, familiarity with YAML/JSON

Learning Outcomes: By the end of this course, students will be able to design persona-based agent workflows, implement governance policies, run AI-mediated simulations, and evaluate cross-project interoperability patterns.

12-Week Syllabus¶

Week 1: Introduction to Agent-Based Workflow Design¶

Session Plan: - Lecture: Evolution from static scripts to agent-based workflows (45 min) - Demo: Install FCC, run fcc --help, explore the CLI dashboard (30 min) - Discussion: When are agent workflows appropriate vs. traditional pipelines? (15 min)

Learning Objectives: - Explain the Find-Create-Critique cycle and its rationale - Install and configure the FCC framework - Navigate the FCC project structure

Reading: Guidebook Chapter 1 (Introduction to FCC)

Week 2: Persona Design and R.I.S.C.E.A.R.¶

Session Plan: - Lecture: The 10-component R.I.S.C.E.A.R. specification (45 min) - Lab: Create a custom persona YAML file (Lab 01) (45 min)

Learning Objectives: - Define all 10 R.I.S.C.E.A.R. components for a persona - Validate persona YAML against the schema - Load personas into the PersonaRegistry

Reading: Guidebook Chapter 2 (Personas)

Week 3: Persona Dimensions and Cross-References¶

Session Plan: - Lecture: 9 dimension categories, 56 attributes, Discernment Matrix (30 min) - Lab: Create dimension profiles and validate against the registry (Lab 03) (30 min) - Exercise: Design a cross-reference entry linking two personas (30 min)

Learning Objectives: - Create dimension profiles with meaningful attribute values - Query the cross-reference matrix for persona interactions - Explain upstream, downstream, and peer relationships

Reading: Guidebook Chapter 3 (R.I.S.C.E.A.R. Dimensions)

Week 4: Workflow Graphs and Actions¶

Session Plan: - Lecture: Workflow graph models, 6 action types, action engine (45 min) - Lab: Walk through the base 5-node workflow graph (Lab 02) (45 min)

Learning Objectives: - Load and traverse workflow graphs - Identify Find, Create, and Critique phases in a graph - Map personas to workflow actions

Reading: Guidebook Chapter 4 (Workflow System)

Week 5: Event-Driven Architecture¶

Session Plan: - Lecture: EventBus pub/sub, 81 event types, filtering, replay (45 min) - Lab: Event-driven simulation (Lab 04) (45 min)

Learning Objectives: - Configure EventBus subscriptions with filters - Capture and inspect simulation events - Replay event sequences for post-hoc analysis

Reading: Guidebook Chapter 6 (Simulation Engine)

Week 6: Plugin System and Extensibility¶

Session Plan: - Lecture: 10 plugin types, cross-plugin orchestration (30 min) - Lab: Develop a persona plugin (Lab 05) (45 min) - Discussion: Plugin architecture trade-offs (15 min)

Learning Objectives: - Implement a custom plugin class - Register plugins and merge personas into the main registry - Explain the plugin type taxonomy

Reading: Guidebook Chapter 5 (Workflow Actions)

Week 7: Governance and Constitutions¶

Session Plan: - Lecture: 3-tier constitutions (hard-stop, mandatory, preferred), quality gates (45 min) - Lab: Governance setup (Lab 08) (45 min)

Learning Objectives: - Define constitutions with rules at each tier - Validate quality gates against expected thresholds - Explain when to use each governance tier

Reading: Guidebook Chapter 9 (Governance)

Week 8: Collaboration Engine and Human-in-the-Loop¶

Session Plan: - Lecture: Session lifecycle, scoring engine, approval gates, handoff protocol (45 min) - Lab: Collaboration session (Lab 09) (45 min)

Learning Objectives: - Create and manage collaboration sessions - Configure approval gates and handoff protocols - Evaluate deliverables using the scoring engine

Reading: Guidebook Chapter 8 (Collaboration Engine)

Week 9: Knowledge Graphs and Semantic Search¶

Session Plan: - Lecture: KnowledgeGraph (9 node types, 9 edge types), builder functions (30 min) - Lecture: Semantic search with embedding providers (30 min) - Lab: Build a knowledge graph and search index (Labs 14-15) (30 min)

Learning Objectives: - Construct knowledge graphs from persona registries - Export graphs to OWL/RDF/SKOS/JSON-LD formats - Build and query persona search indexes

Reading: Guidebook Chapter 11 (Phase 13 Content)

Week 10: RAG Pipeline and Protocol Integration¶

Session Plan: - Lecture: DocumentChunker strategies, SemanticRetriever, RAGPipeline (30 min) - Lecture: A2A Agent Cards, MCP Server, Protocol Bridge (30 min) - Lab: RAG pipeline from scratch (Lab 16) and Protocol Bridge (Lab 17) (30 min)

Learning Objectives: - Build an end-to-end RAG pipeline with persona-aware queries - Generate A2A Agent Cards from persona specifications - Route protocol messages through the bridge

Reading: API Reference (RAG, Protocols)

Week 11: Federation and Cross-Project Integration¶

Session Plan: - Lecture: EntityResolver, NamespaceRegistry, FederatedKnowledgeGraph (30 min) - Lab: Federated knowledge query (Lab 18) (30 min) - Guest speaker: Cross-organizational ontology alignment challenges (30 min)

Learning Objectives: - Resolve entities across federated namespaces - Build federated knowledge graphs with cross-namespace edges - Assess cross-project alignment quality

Reading: API Reference (Federation, Knowledge Graphs)

Week 12: Capstone Presentations and Peer Review¶

Session Plan: - Student presentations: 10-minute capstone demos (60 min) - Peer review: Written feedback using structured rubric (30 min)

Learning Objectives: - Demonstrate mastery of all major FCC subsystems - Provide constructive peer feedback - Reflect on design decisions and trade-offs

Grading Rubric¶

Component	Weight	Description
Lab Exercises (10 labs)	30%	Completion and quality of lab deliverables
Knowledge Checks	10%	15 multiple-choice assessments
Exercises (7 written)	20%	Depth of analysis and design quality
Capstone Project	30%	Full-pipeline FCC extension (see assessment_rubrics.yaml)
Participation & Peer Review	10%	Discussion engagement and peer feedback quality

Guest Speaker Suggestions¶

Topic	Suggested Background
Multi-agent system design in production	Industry practitioner (platform engineering)
AI governance and responsible AI	Ethics researcher or policy specialist
Ontology alignment and federation	Knowledge engineering or semantic web researcher
Open-source framework maintenance	Open-source project maintainer

Research Project Guidelines¶

Students may propose a research project as an alternative to the capstone. Requirements:

Hypothesis: A falsifiable claim about FCC-mediated workflows (e.g., "Adding a governance gate improves output quality by X%")
Methodology: Use FCC simulation traces as data; define independent and dependent variables
Implementation: At least 3 custom personas, 1 custom workflow, reproducible simulation runs
Analysis: Statistical analysis of simulation traces using the scoring engine
Report: 8-12 pages, IEEE or ACM format, including related work section

Peer Review Process¶

For each capstone presentation:

Each reviewer completes a structured rubric (see assessment_rubrics.yaml peer review section)
Reviews are anonymous unless the class agrees otherwise
Reviewers assess: technical correctness, design justification, code quality, presentation clarity
Authors receive consolidated feedback within 48 hours
Authors may submit a 1-page response addressing reviewer comments

Course Materials¶

All course materials are included in the FCC package:

Resource	Location
Guidebook (18 chapters)	`docs/guidebook/`
Lab exercises (18 labs)	`src/fcc/data/docs/lab_exercises.yaml`
Assessment rubrics	`src/fcc/data/docs/assessment_rubrics.yaml`
Jupyter notebooks (18)	`notebooks/`
Streamlit apps (20)	`apps/streamlit/`
Book series (3 books)	`docs/books/`