Demo: Team Knowledge Onboarding¶

A new developer joins your team. Instead of spending 2 weeks reading docs and asking questions, their AI agent understands the project from minute one.

Scenario Setup¶

Your team has a payment processing service. Senior developers have been working on it for months with fcontext enabled. The .fcontext/ directory is committed to Git.

.fcontext/
├── _README.md                  # "This is a payment gateway service using 
│                               #  Stripe API with idempotency keys..."
├── _workspace.map              # Project structure overview
├── _cache/
│   ├── docs/
│   │   └── stripe-integration.pdf.md    # Indexed Stripe contract
│   └── specs/
│       └── pci-compliance.docx.md       # PCI requirements
├── _topics/
│   ├── idempotency-pattern.md           # Why we use idempotency keys
│   ├── retry-strategy.md               # Exponential backoff decisions
│   └── currency-rounding.md            # Currency precision rules
├── _requirements/
│   ├── EPIC-001.yaml                    # Payment flow
│   ├── STORY-003.yaml                  # Refund handling
│   └── TASK-012.yaml                   # Add JPY support (in-progress)
└── _experiences/
    └── payment-domain/                  # Imported: payment industry knowledge
        ├── _README.md
        ├── _cache/
        └── _topics/

New Developer — Day 1¶

Alice joins the team. She clones the repo:

git clone git@github.com:your-org/payment-service.git
cd payment-service
fcontext enable cursor    # Alice uses Cursor

Alice opens Cursor and asks:

"What is this project and how is it structured?"

Cursor reads .fcontext/_README.md and _workspace.map, and responds with a precise overview — architecture, key modules, API patterns, and current work items. No senior dev needed for the intro.

Day 1 — Understanding Domain Rules¶

Alice asks:

"Why do we use idempotency keys on every payment request?"

Cursor reads .fcontext/_topics/idempotency-pattern.md and explains the exact reasoning — network retries, duplicate charge prevention, the specific Stripe API behavior. Knowledge that took the team weeks to formalize is available instantly.

Day 1 — Reading Specs¶

Alice asks:

"What are our PCI compliance requirements?"

Cursor reads .fcontext/_cache/specs/pci-compliance.docx.md — the original DOCX was indexed by a senior dev months ago. Alice gets the full spec without hunting for the document.

Day 2 — First Task¶

Alice picks up TASK-012 (Add JPY support). She asks:

"What's the current status of JPY support and what do I need to know?"

Cursor reads:

_requirements/TASK-012.yaml — task details, parent story, acceptance criteria
_topics/currency-rounding.md — why JPY has no decimal places (¥100, not ¥1.00)
_experiences/payment-domain/ — general payment industry rules about currency handling

Alice writes the implementation. Cursor guides her based on existing patterns and domain knowledge. She's productive on Day 2, not Week 2.

How the Team Built This Context¶

It wasn't a one-time effort. The context accumulated naturally over months:

graph TD subgraph "Month 1" M1A["fcontext init"] M1B["Index PCI spec"] M1C["First topics written"] end subgraph "Month 2" M2A["More topics accumulated"] M2B["Requirements added"] M2C["Experience pack imported"] end subgraph "Month 3" M3A["Alice joins"] M3B["Instant full context"] M3C["Productive Day 2"] end M1A --> M1B --> M1C --> M2A --> M2B --> M2C --> M3A --> M3B --> M3C

Each developer's AI sessions added knowledge to .fcontext/. By the time Alice joined, the context was rich and comprehensive — not because someone wrote documentation, but because the AI captured knowledge as a side effect of daily work.

Try It Yourself¶

# Senior dev: set up context
cd your-project
fcontext init
fcontext enable copilot
fcontext index docs/        # Index existing documents

# Work normally — AI accumulates knowledge in _topics/

# When ready to share:
git add .fcontext/
git commit -m "project context"
git push

# New member:
git clone your-project.git
fcontext enable cursor     # Their preferred agent
# Done — instant onboarding