Skip to main content

FCC (Find → Create → Critique) Agent Team Framework

Project description

FCC Agent Team Extension

PyPI version Python 3.10+ License: MIT Tests: 1168 Coverage: 100%

FCC (Find → Create → Critique) agent workflow framework — a clean, well-structured Python package for orchestrating documentation agent teams.

Quick Start

For Users

pip install fcc-agent-team-ext
fcc --version

For Contributors

git clone https://github.com/rollingthunderfourtytwo-afk/l2_fcc_agent_team_ext.git
cd l2_fcc_agent_team_ext
make venv && source .venv/bin/activate
make install-dev
make test && make lint

What is FCC?

FCC is a three-phase workflow for documentation creation:

  1. Find — Research personas gather and synthesize information
  2. Create — Design, operations, and user-facing personas produce artifacts
  3. Critique — Review and governance personas ensure quality

24 personas across 5 categories collaborate through message-passing workflows with feedback loops:

Category Count Personas
Core 5 Research Crafter, Blueprint Crafter, Documentation Evangelist, Runbook Crafter, User Guide Crafter
Integration 7 Catalog Indexer Architect, UI Mockup Crafter, Semantic Taxonomy Engineer, Traceability Specialist, Blueprint Validator, Research Inventory Crafter, Governance Compliance Auditor
Governance 3 Data Governance Specialist, Privacy Taxonomy Engineer, Anti-fact Mitigation Specialist
Stakeholder 5 Collaboration Orchestrator, SAFe Metrics Crafter, Executive Communicator, Roadmap Synchronizer, Stakeholder Content Publisher
Champion 4 Research Crafter Champion, Blueprint Crafter Champion, User Guide Crafter Champion, Runbook Crafter Champion

Champions are elevated personas that orchestrate teams of base personas (e.g., Research Crafter Champion coordinates RC, CIA, STE, and RIC).

Who Is This For?

  • Beginners / Students — Learn agent orchestration patterns, persona-based workflows, and documentation-as-code methodologies through a well-structured, tested codebase with comprehensive examples.
  • Developers — Build agent-powered documentation pipelines using a modular Python framework with CLI tooling, simulation engines, and Jinja2 template generation.
  • Researchers — Study the R.I.S.C.E.A.R. persona specification, 56-dimension persona profiling, Discernment Matrix, and Design Target Factors as formal models for human-AI collaboration.
  • Practitioners — Deploy FCC workflows in enterprise settings with governance controls, quality gates, capability tagging, and Scaled Agile alignment.

R.I.S.C.E.A.R. Framework

The full R.I.S.C.E.A.R. specification defines personas with 10 core components:

Component Purpose
Role Identity and function
Input Data and background information
Style Tone, language, and format
Constraints Limitations and requirements
Expected Output Desired structure and detail
Archetype Fundamental character/persona
Responsibilities Duties and ethical obligations
Role Skills Skills for job execution
Role Collaborators Team collaboration patterns
Role Adoption Checklist Integration validation

Plus: Discernment Matrix (6 traits), Design Target Factors (6 factors), and a 56-dimension Persona system across 9 categories — all 24 personas fully populated with dimension profiles, discernment matrices, and design target factors.

Key Concepts

  • FCC Cycle — The three-phase workflow (Find → Create → Critique) that structures how documentation is researched, produced, and reviewed. Artifacts flow forward through phases, with feedback edges enabling iterative refinement.
  • Personas — AI agent roles defined by the R.I.S.C.E.A.R. specification. Each persona has a specific function, communication style, constraints, and collaboration patterns. 24 personas span 5 categories.
  • Champions — Elevated personas that orchestrate teams of base personas. Each champion coordinates 3-4 team members, enabling top-down coordination alongside bottom-up message passing.
  • Cross-Reference Matrix — A queryable 100+ entry interaction map across 5 relationship types (handoff, feedback, coordination, governance, champion-of). Supports upstream, downstream, peer, and relationship-type queries to understand how all 24 personas collaborate.
  • Workflow Graphs — Directed graphs where nodes are personas and edges are message-passing connections. Three predefined graphs (5-node, 20-node, 24-node) support different levels of complexity. Traversed via BFS for deterministic activation ordering.
  • Scenarios — Predefined test cases that exercise specific FCC workflows. Each scenario specifies participating personas, the workflow graph, and validation rules. 17 scenarios cover basic through advanced orchestration patterns.
  • Quality Gates — Automated validation checkpoints that enforce documentation standards. 28 gates with configurable severity (error, warning, info) ensure completeness, compliance, and structural integrity. Includes 3 cross-reference integrity gates.

Project Structure

src/fcc/
├── _resources.py    Centralized data/template path resolution (importlib.resources)
├── personas/        Persona models, R.I.S.C.E.A.R. specs, dimensions, registry, cross-reference matrix
├── workflow/        Workflow graph models, traversal, BFS
├── simulation/      Deterministic + AI-powered simulation engine
├── scenarios/       Scenario loading, dynamic validation framework
├── scaffold/        CLI tool (fcc init, add-persona, validate, generate-docs, sitemap)
├── governance/      Capability tags, quality gates, compliance
├── data/            Package data (included in wheel distribution)
│   ├── personas/    24 personas across 5 YAML files + cross-reference matrix + dimensions
│   ├── workflows/   3 workflow graphs (base 5-node, extended 20-node, complete 24-node)
│   ├── schemas/     JSON Schema for personas, workflows, traces, cross-references
│   ├── scenarios/   17 scenarios (6 starter + 11 advanced)
│   ├── governance/  Tag registry (30 tags) + quality gate definitions (28 gates)
│   └── docs/        Topic registries (21 tutorials, 24 prompts, constitution template)
└── templates/
    ├── project/     Jinja2 templates for project scaffolding
    └── docs/        15 Jinja2 templates for docs-as-code generation

tests/               1168 tests across 24 test files (100% coverage)

Architecture Overview

graph TB
    YAML[YAML Personas] --> REG[PersonaRegistry]
    REG --> SIM[Simulation Engine]
    REG --> DOC[Doc Generator]
    SIM --> TRACE[Traces]
    DOC --> DOCS[1,348 Doc Files]
    GOV[Governance] --> SIM
    CLI[fcc CLI] --> SIM
    CLI --> DOC

CLI

fcc init --name "my-project" --dir ./my-project          # Bootstrap FCC project
fcc add-persona "Custom Analyst" --phase Find --id CA    # Add persona
fcc validate                                              # Check project structure
fcc simulate --scenario GEN-001                           # Run simulation
fcc generate-docs --dir ./docs --personas all             # Generate all 1,348 doc files
fcc generate-docs --dir ./docs --personas core            # Generate core persona docs only
fcc validate-docs --dir ./docs                            # Validate generated docs
fcc sitemap --dir ./docs                                  # Generate SITEMAP.md

Phase Roadmap

Phase Status Tests Description
1. Foundation Done 3 Project skeleton, packaging, smoke tests
2. Core FCC Done 43 Personas, workflow graph, registry
3. Simulation Done 32 Engine, traces, AI client, scenarios
4. Scaffolding CLI Done 22 fcc init, templates, project config
5. Governance Done 21 Tags, quality gates, compliance
R.I.S.C.E.A.R. Expansion Done 259 total Full 10-component spec, dimensions, strategic architecture
Extended Persona Gallery Done 693 total 24 personas, cross-reference matrix, champion architecture, docs-as-code (1,348 files), dynamic validation
CI/CD, Scenarios & Docs Done 726 total GitHub Actions CI/CD, MkDocs deployment, 11 advanced scenarios, 5 new validation rules, comprehensive documentation
Coverage Push to 100% Done 757 total 100% line coverage across 20 test files
Phase 5.5: Profiles & Docs UX Done 877+ total 24 dimension profiles, documentation overhaul, MkDocs UX, 50 hand-written pages, 13 tutorials
Phase 6: Cross-Reference Expansion Done 1030+ total Cross-reference matrix 16→106 entries, governance/champion-of types, 3 quality gates, 9 new doc pages
Phase 7: Distribution Done 1168 total PyPI packaging, release automation, data relocation, _resources module, comprehensive distribution docs

Documentation Site

The FCC documentation site is built with MkDocs Material and offers 8 audience-targeted navigation paths:

Audience Entry Point
Beginners / Students Getting Started → Tutorials
Developers Developer Guide
Researchers Specification
Practitioners Tutorials → Customization
Decision Makers For Decision Makers
Managers / Team Leads For Teams
Content Strategists Personas → Specification
Open Source Contributors Developer Guide → Contributing

Next Steps

  • Advanced Automation (Phase 8) — Custom persona templates, plugin system, template marketplace
  • Agent-Human Collaboration (Phase 9) — Feedback loop instrumentation, human-in-the-loop handoff protocols
  • Scaled Agile Integration (Phase 10) — Architectural runway management, portfolio-level orchestration

Documentation

Contributing

Contributions are welcome! Whether you are fixing a bug, adding a persona, improving documentation, or proposing a new feature, we would appreciate your help. Please read the CONTRIBUTING.md guide for development setup, code style, testing requirements, and the pull request process.

License

MIT License. Copyright (c) INFORMATION COLLECTIVE, LLC. See LICENSE for details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fcc_agent_team_ext-0.2.0.tar.gz (182.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

fcc_agent_team_ext-0.2.0-py3-none-any.whl (225.4 kB view details)

Uploaded Python 3

File details

Details for the file fcc_agent_team_ext-0.2.0.tar.gz.

File metadata

  • Download URL: fcc_agent_team_ext-0.2.0.tar.gz
  • Upload date:
  • Size: 182.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for fcc_agent_team_ext-0.2.0.tar.gz
Algorithm Hash digest
SHA256 8ee0f4872a3fdfedb373cbaac780621ebed642742f7f763d66f6545ca9ffdc30
MD5 365a4a55cbf9baf0094b9787e718dc81
BLAKE2b-256 6bae27f10432b5359c8460036e2e6696a024b63ac2b3830f867dd91c6fae27dd

See more details on using hashes here.

Provenance

The following attestation bundles were made for fcc_agent_team_ext-0.2.0.tar.gz:

Publisher: release.yml on rollingthunderfourtytwo-afk/l2_fcc_agent_team_ext

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file fcc_agent_team_ext-0.2.0-py3-none-any.whl.

File metadata

File hashes

Hashes for fcc_agent_team_ext-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 b8dc1f0f7bd20ce5fffe22335b213ed53d1ee1af1088a24527ffa810be4a8dcf
MD5 acb51254d19cf6ae65516394af6e1298
BLAKE2b-256 95aea48295bd84f0cf1cf53c1986720a4852cd00af1879aa10936bf28848ded8

See more details on using hashes here.

Provenance

The following attestation bundles were made for fcc_agent_team_ext-0.2.0-py3-none-any.whl:

Publisher: release.yml on rollingthunderfourtytwo-afk/l2_fcc_agent_team_ext

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page