Skip to content

setup.md

Latest commit

 

History

History
148 lines (113 loc) · 4.23 KB

setup.md

File metadata and controls

148 lines (113 loc) · 4.23 KB
Raw

TestPilot AI — Developer Setup Guide

This guide covers all options for running TestPilot AI locally, including environment configuration, service endpoints, and Makefile automation.

Option 1: 60-Second Containerized Deployment (Recommended)

git clone https://github.com/Shikhaar/TestPilot-AI.git
cd TestPilot-AI
cp .env.example .env
docker compose up -d
docker compose exec backend alembic upgrade head

Open http://localhost:3000.

Option 2: Native Host Development Setup

Prerequisites

  • Python 3.12 & Poetry
  • Node.js v18+ & npm
  • Docker & Docker Compose (for PostgreSQL, Redis, and Qdrant)

1. Start Infrastructure Services

docker compose up -d postgres redis qdrant

2. Configure Host Connection Strings (.env)

When running the backend natively, replace the Docker service hostnames (postgres, redis, qdrant) with localhost. All other values stay the same as .env.example.

# Application
APP_NAME=TestPilot AI
APP_ENV=development
DEBUG=true
SECRET_KEY=change-me-to-a-very-long-random-secret-key-in-production
ALLOWED_ORIGINS=http://localhost:3000,http://localhost:80

# PostgreSQL (use localhost instead of Docker service name)
POSTGRES_HOST=localhost
POSTGRES_PORT=5432
POSTGRES_USER=testpilot
POSTGRES_PASSWORD=testpilot_secret
POSTGRES_DB=testpilot
DATABASE_URL=postgresql+asyncpg://testpilot:testpilot_secret@localhost:5432/testpilot

# Redis (use localhost instead of Docker service name)
REDIS_HOST=localhost
REDIS_PORT=6379
REDIS_PASSWORD=redis_secret
REDIS_URL=redis://:redis_secret@localhost:6379/0
CELERY_BROKER_URL=redis://:redis_secret@localhost:6379/0
CELERY_RESULT_BACKEND=redis://:redis_secret@localhost:6379/1

# Qdrant (use localhost instead of Docker service name)
QDRANT_HOST=localhost
QDRANT_PORT=6333
QDRANT_API_KEY=qdrant_secret
QDRANT_URL=http://localhost:6333

# GitHub App
GITHUB_APP_ID=
GITHUB_APP_PRIVATE_KEY_PATH=/secrets/github-private-key.pem
GITHUB_WEBHOOK_SECRET=your-webhook-secret
GITHUB_CLIENT_ID=
GITHUB_CLIENT_SECRET=
GITHUB_APP_NAME=testpilot-ai-app

# JWT Authentication
JWT_SECRET_KEY=change-me-to-another-long-random-key
JWT_ALGORITHM=HS256
JWT_ACCESS_TOKEN_EXPIRE_MINUTES=60
JWT_REFRESH_TOKEN_EXPIRE_DAYS=30

# LiteLLM / OpenAI
OPENAI_API_KEY=sk-placeholder-replace-with-real-key
LITELLM_DEFAULT_MODEL=gpt-4o-mini
LITELLM_FALLBACK_MODEL=ollama/mistral
USE_LOCAL_EMBEDDINGS=true
SENTENCE_TRANSFORMER_MODEL=all-MiniLM-L6-v2

# Frontend
NEXT_PUBLIC_API_URL=http://localhost:8000/api/v1
NEXT_PUBLIC_WS_URL=ws://localhost:8000/ws
NEXT_PUBLIC_APP_NAME=TestPilot AI

3. Start Backend API

cd backend
poetry install
poetry run alembic upgrade head
poetry run uvicorn app.main:app --port 8000 --reload

4. Start Frontend Client

cd frontend
npm install
npm run dev

Open http://localhost:3000.

Service Endpoints

Service Endpoint Purpose
Web Interface http://localhost:3000 Application frontend dashboard
Backend Swagger API http://localhost:8000/docs Interactive OpenAPI specification
Qdrant Vector Dashboard http://localhost:6333/dashboard Vector storage collection manager
Celery Flower http://localhost:5555 Distributed task execution monitor
Grafana Dashboard http://localhost:3001 Prometheus telemetry visualization

Makefile Automation

Command Action
make dev Launch complete stack via Docker Compose
make stop Terminate all active container services
make migrate Execute pending Alembic migrations
make db-reset Reset database state and re-apply schema
make lint Run Ruff static analysis across Python modules
make test Execute Pytest automated test suite

Usage Workflow

  1. Access the Dashboard at http://localhost:3000.
  2. Connect a Repository — select from your linked GitHub account or enter any public repository URL.
  3. Observe AST Indexing — track real-time AST parsing progress, health scores, and coverage metrics.
  4. Code Search & Unit Testing — execute 3-layer hybrid code searches or generate automated unit test suites.
  5. Disconnect a Repository — purge storage directories and Qdrant vector collections with one click.