Cisco AI Agent – Mono‑Repo

Purpose  Bootstrap directory structure, local dev workflow, and basic Git commands.

---

1  Folder Layout

repo/
├─ client/          # React PWA
├─ api/               # FastAPI + LangGraph orchestrator
├─ infra/             # Terraform or CDK modules
└─ scripts/           # Utility scripts (index build, data ingest)

---

2  Prerequisites

Tool	Minimum Version	Purpose
Git	2.40	version control
Docker	24.x	local containers
Python	3.11	API & tooling
Node + PNPM	20.x + 8.x	React front‑end
Make (optional)	any	convenience tasks

---

3  Clone & Bootstrap

# 1. Clone and enter repo
$ git git@github.com:jafetmora/bd-cisco-poc.git
$ cd bd-cisco-poc

# 2. Install pre‑commit hooks (lint/format)
$ pip install pre-commit && pre-commit install

# 3. Build & run both services with Docker Compose
$ docker compose up --build

Local endpoints:

• FastAPI – http://localhost:8000/docs
• React   – http://localhost:5173

---

4  Development Branching

main   ← protected, demo deploys
 dev    ← integration, staging deploys
 feature/*  ← short-lived work branches
 infra/*    ← Infra changes only

Create a feature branch and push:

$ git checkout -b feature/my-task
# …code…
$ git commit -m "feat: my task"
$ git push -u origin feature/my-task

Open a Pull Request into test. GitHub Actions will lint, test, and scan images.

---

5  Docker Quick‑start

Build images locally

# API
$ docker build -t ai-agent-api:dev -f api/Dockerfile .

# Front‑end
$ docker build -t ai-agent-client:dev -f client/Dockerfile .

Run with Compose

$ docker compose up

---

6  Makefile Targets (optional)

make dev        # compose up (live reload)
make lint       # ruff + eslint
make test       # pytest + vitest
make clean      # docker prune + cache clear