Skip to content

Repository files navigation

Testing AI Assistant

Full documentation available in docs/en/ | docs/ru/

Testing AI Assistant is an AI-powered test orchestration platform built on a microservices architecture. It automates test execution pipelines, generates tests using LLM agents, detects flaky tests and bugs, and delivers real-time notifications across multiple channels — all managed through a web dashboard and mobile app.

Key capabilities:

  • Run existing tests from repositories (unit, lint, SAST, E2E, load, coverage)
  • AI-generated test checklists from app description or URL
  • AI-generated Playwright E2E tests from checklist items
  • Execute tests against a live application with real-time results
  • Import/export test checklists as JSON

Technology Stack

Layer Technology
Frontend Next.js 15, React 19, Tailwind CSS, shadcn/ui, next-intl
Mobile Expo (React Native), Expo Router
Backend NestJS (Node.js), TypeScript
Databases PostgreSQL (per service), Redis (caching / queues)
Messaging Redpanda (Kafka-compatible)
Workflow Temporal
AI Claude API (Anthropic)
Storage MinIO (S3-compatible)
Observability OpenTelemetry, Prometheus, Grafana, Loki, Tempo
Infrastructure Docker Compose (dev), Kubernetes + Helm (prod), Terraform (AWS)
CI/CD GitHub Actions, ArgoCD

Architecture

┌─────────────┐  ┌──────────────┐
│  Dashboard   │  │  Mobile App  │
│  (Next.js)   │  │  (Expo)      │
└──────┬───────┘  └──────┬───────┘
       │                 │
       └────────┬────────┘
                │
     ┌──────────▼──────────┐
     │   API Gateway /     │
     │   Traefik           │
     └──────────┬──────────┘
                │
  ┌─────────────┼─────────────────────────────┐
  │             │             │               │
  ▼             ▼             ▼               ▼
Identity    Organization   Project        Pipeline
Service     Service        Service        Service
  │             │             │               │
  ▼             ▼             ▼               ▼
pg-identity  pg-org       pg-project     pg-pipeline
                                              │
                                    ┌─────────▼─────────┐
                                    │  Temporal Server   │
                                    └─────────┬─────────┘
                                              │
                                    ┌─────────▼─────────┐
                                    │   Test Runner      │
                                    │   (Temporal Worker) │
                                    └─────────┬─────────┘
                                              │
                              ┌───────────────┼───────────────┐
                              ▼               ▼               ▼
                         AI Service    Notification      MinIO
                         (Claude)      Service           (Artifacts)

Microservices

Service Port Description
identity 3001 Authentication, JWT, OAuth (GitHub/GitLab/Bitbucket)
organization 3002 Organizations and membership management
project 3003 Project and git repository management
pipeline 3004 Pipeline definitions, test runs, results, SSE
ai 3005 LLM agents: test generation, bug detection, flaky tests
notification 3006 Email, Slack, Telegram, push notifications
test-runner Temporal worker: executes pipeline steps (unit, lint, SAST, E2E, etc.)

Quick Start

# Prerequisites: Node.js 20+, pnpm 9+, Docker

# 1. Clone and install
git clone https://github.com/micode-ai/testing-ai-assistant.git
cd testing-ai-assistant
pnpm install

# 2. Start infrastructure (PostgreSQL, Redis, Temporal, MinIO, etc.)
docker compose up -d

# 3. Copy environment files
cp .env.example .env
cp apps/dashboard/.env.example apps/dashboard/.env.local

# 4. Run database migrations
pnpm -r exec prisma migrate dev

# 5. Start all services in dev mode
pnpm dev

Dashboard will be available at http://localhost:3000.

Documentation

License

Private — all rights reserved.

About

Testing AI Assistant is an AI-powered platform that analyzes your codebase to detect bugs, find flaky tests, and generate production-ready test files automatically. Connected to GitHub, GitLab, and Bitbucket, it runs test pipelines and delivers results via Email, Slack, and Telegram through a single dashboard.

Resources

Stars

4 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages