iavipro/testteller

By iavipro

•Updated over 1 year ago

AI agent to generate contextual test cases & automation based on project docs & code.

Image

Developer tools

1.1K

Overview Tags

iavipro/testteller repository overview

⁠TestTeller Agent

An AI-powered test-case generation and automation tool that leverages RAG architecture to analyze project documentation (PRDs, Design Docs, Contracts etc.) and project code to generate comprehensive test cases and then effectively automate them, using multiple supported languages and frameworks, while leveraging GenAI / LLMs ( Gemini, Ollama, Claude, OpenAI )

⁠🚀 Features

Multi-LLM Support: Works with Gemini, OpenAI, Claude, and Llama/Ollama
Document Processing: Supports multiple formats (PDF, DOCX, Markdown, Excel)
Vector Database: ChromaDB integration for intelligent document retrieval
Test Automation: Generates executable test code in multiple frameworks
Docker-First: Production-ready containerized deployment
Flexible Configuration: Environment-based configuration for different setups

⁠📋 Prerequisites

Docker and Docker Compose
API keys for your chosen LLM provider
(Optional) Ollama for local ( or within accessible network ) LLM inference

⁠🔧 Quick Start

⁠1. Create Environment File

# Create .env file with your configuration
cat > .env << 'EOF'
# Choose your LLM provider: gemini, openai, claude, llama
LLM_PROVIDER=gemini

# API Keys (configure based on your provider)
GOOGLE_API_KEY=your_gemini_api_key_here
OPENAI_API_KEY=your_openai_api_key_here
CLAUDE_API_KEY=your_claude_api_key_here

# ChromaDB Configuration
CHROMA_DB_HOST=chromadb
CHROMA_DB_PORT=8000
CHROMA_DB_USE_REMOTE=true
DEFAULT_COLLECTION_NAME=test_documents

# Logging
LOG_LEVEL=Error
LOG_FORMAT=json
EOF

⁠2. Start Services

# Using Docker Compose
docker-compose up -d

# Or using Docker Run (requires separate ChromaDB)
docker run -d \
  --name testteller \
  --env-file .env \
  -v testteller_data:/app/chroma_data \
  iavipro/testteller:latest

⁠3. Generate Test Cases

# Help
docker-compose exec app testteller -h

# Configure
docker-compose exec app testteller configure

# Ingest documents
docker-compose exec app testteller ingest-docs /path/to/your/documents --collection-name your_collection_name

# Ingest code
docker-compose exec app testteller ingest-code /path/to/your/code --collection-name your_collection_name

# Generate test cases
docker-compose exec app testteller generate "Give more than 20 technical test cases" --collection-name your_collection_name -o test.md

# Automate test generation
docker-compose exec app testteller automate /your/generated/test-cases/file --language python --framework pytest --collection-name your_collection_name

⁠🔨 Usage Examples

⁠Document Ingestion

# Ingest local documents
docker-compose exec app testteller ingest-docs /app/local_docs -c your_collection_name

# Ingest from Git repository
docker-compose exec app testteller ingest-code https://github.com/user/repo.git -c your_collection_name

⁠Test Generation

# Generate test cases for specific files
docker-compose exec app testteller generate "All Integration tests" -c your_collection_name -f pdf

# Generate with custom output
docker-compose exec app testteller generate "All Integration tests" -c your_collection_name --output-file custom_tests.md

⁠Test Automation

# Python/pytest automation
docker-compose exec app testteller automate /your/generated/test-cases/file --language python --framework pytest -c your_collection_name

# JavaScript/Jest automation
docker-compose exec app testteller automate /your/generated/test-cases/file --language javascript --framework jest -c your_collection_name

⁠⚙️ Configuration

⁠Environment Variables

Variable	Description	Default
`LLM_PROVIDER`	LLM provider (gemini/openai/claude/llama)	`gemini`
`GOOGLE_API_KEY`	Google Gemini API key	-
`OPENAI_API_KEY`	OpenAI API key	-
`CLAUDE_API_KEY`	Anthropic Claude API key	-
`OLLAMA_BASE_URL`	Ollama server URL	`http://host.docker.internal:11434`
`CHROMA_DB_HOST`	ChromaDB host	`chromadb`
`CHROMA_DB_PORT`	ChromaDB port	`8000`
`CHUNK_SIZE`	Document chunk size	`1000`
`CHUNK_OVERLAP`	Document chunk overlap	`200`
`LOG_LEVEL`	Logging level	`INFO`

⁠Volume Mounts

volumes:
  # Persistent ChromaDB data
  - chroma_data:/app/chroma_data
  
  # Temporary repository clones
  - ./temp_cloned_repos:/app/temp_cloned_repos
  
  # Optional: Local documents
  - ./local_docs:/app/local_docs:ro

⁠🌐 Deployment Scenarios

⁠Local Development

# Use with local Ollama
LLM_PROVIDER=llama
OLLAMA_BASE_URL=http://host.docker.internal:11434

⁠Remote ChromaDB

# Connect to external ChromaDB
CHROMA_DB_HOST=your-chromadb-server.com
CHROMA_DB_PORT=8000
CHROMA_DB_USE_REMOTE=true

⁠Production Deployment

# Use managed LLM services
LLM_PROVIDER=gemini
GOOGLE_API_KEY=your_production_api_key
LOG_LEVEL=WARNING

⁠🔍 Health Checks

The container includes health checks for:

ChromaDB connectivity
LLM provider availability
Application responsiveness

# Check service health
docker-compose ps
docker-compose logs app

⁠📊 Resource Requirements

⁠Minimum Requirements

CPU: 0.5 cores
Memory: 1GB
Storage: 2GB

⁠Recommended Production

CPU: 2 cores
Memory: 4GB
Storage: 10GB

⁠🐛 Troubleshooting

⁠Common Issues

ChromaDB Connection Failed

# Check ChromaDB health
docker-compose exec chromadb curl -f http://localhost:8000/api/v1/heartbeat

Ollama Connection Refused

# Ensure Ollama is running on host
curl http://localhost:11434/api/health

# For remote Ollama, update OLLAMA_BASE_URL
OLLAMA_BASE_URL=http://your-ollama-server:11434

⁠📖 Documentation

⁠🤝 Support

🐛 Issues: GitHub Issues⁠

⁠📄 License

This project is licensed under the Apache License 2.0 - see the LICENSE⁠ file for details.

Tag summary

Recent tags

Content type

Image

Digest

sha256:7579d6c95…

Size

299.4 MB

Last updated

over 1 year ago

docker pull iavipro/testteller