Skip to content

Quickstart

This guide takes you from zero to your first FCC simulation in under ten minutes. By the end, you will have run a mock simulation, generated documentation files, and explored the output.

Prerequisites

  • Python 3.10 or later
  • pip and venv (included with standard Python installations)
  • A terminal with access to make (optional but recommended)

Step 1: Install

Install from PyPI:

pip install fcc-agent-team-ext

Or for development (editable install with dev dependencies):

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

Step 2: Verify the Installation

Confirm FCC is installed and accessible:

fcc --help

You should see the CLI help output listing available commands: init, add-persona, validate, simulate, generate-docs, validate-docs, sitemap, collab, dashboard, plugins, and action.

Step 3: Run Your First Simulation

Execute a mock simulation using the base workflow and the GEN-001 starter scenario:

fcc simulate --scenario GEN-001

This runs the 5-node base workflow in deterministic mock mode. No API keys are required. The simulation traverses the workflow graph -- Research Crafter, Blueprint Crafter, Documentation Evangelist, Runbook Crafter, User Guide Crafter -- generating a trace at each step.

When complete, you will see output like:

Simulation complete: 5 steps, 5 AI calls
Traces written to: traces_ai.json

Open traces_ai.json to inspect the structured trace. Each event records the step number, acting persona, payload, edge label, and AI response metadata.

Step 4: Generate Documentation

Use the docs-as-code generator to produce documentation files from persona specifications:

fcc generate-docs --dir docs_output

This generates documentation for all 102 core personas using Jinja2 templates. The output includes tutorials, prompts, and workflow files -- up to 56 files per persona.

Validate the generated documentation:

fcc validate-docs --dir docs_output

Step 5: Explore the Output

After generation, explore the output structure:

ls docs_output/fcc/personas/

Each persona has its own directory containing tutorials, prompt templates, and workflow documentation. A sitemap can be generated for navigation:

fcc sitemap --dir docs_output

Step 6: Explore Plugins and Actions

List installed plugins:

fcc plugins list

List available workflow actions for all personas:

fcc action list

List actions for a specific persona:

fcc action list --persona RC

Step 7: View Dashboards

FCC includes terminal-based dashboards for exploring the ecosystem:

# Browse the persona catalog
fcc dashboard personas

# View quality gate status
fcc dashboard quality

# View the ecosystem overview
fcc dashboard ecosystem

Step 8: Start a Collaboration Session

Launch an interactive human-in-the-loop session:

fcc collab start --workflow base_sequence --participants RC,BC,human-reviewer

This creates a collaboration session where AI personas and human reviewers take turns contributing to a deliverable. The session tracks turns, evaluates quality gates, and manages handoff between agents and humans.

What Just Happened

In these steps, you:

  1. Installed the FCC framework as an editable Python package.
  2. Ran a simulation that traversed a 5-node workflow graph with deterministic AI responses.
  3. Generated documentation from 102 core persona specifications using Jinja2 templates.
  4. Validated the generated output for completeness.
  5. Explored the plugin system and available workflow actions.
  6. Viewed the terminal dashboards for ecosystem monitoring.
  7. Started a human-in-the-loop collaboration session.

Next Steps

  • Installation -- Detailed setup including optional dependencies and AI provider configuration
  • Key Concepts -- Understand the vocabulary: personas, workflows, quality gates
  • What is FCC? -- The conceptual foundation behind the framework
  • Value Proposition -- Why FCC matters for documentation programs