Intelligent research and technical specification generator using LangGraph workflows.
- Techspec Workflow: Automated connector code generation with validation and documentation
- LangGraph Integration: State-based workflow orchestration with parallel processing
- Rich Output: Multiple formats (Markdown, JSON, Text) with comprehensive metadata
- Python 3.9+ (Required for LangGraph compatibility)
- uv or pip for package management
# Install from source
cd grace
uv sync
# Or install specific feature groups
uv sync --extra dev --extra ai --extra scraping# Install in development mode
uv pip install -e .
# Or with optional dependencies
uv pip install -e ".[dev,ai,scraping,nlp]"# Generate connector for a payment processor
source .venv/bin/activate # to use grace from outside folders as well
# move the grace/.env.example to .env -> and update Techspec output path and API keys if needed
grace techspecworkflows use LangGraph for sophisticated state management and parallel processing:
API Analysis -> Schema Extract -> Code Generation
|
Finalize Output <- Generate Docs <- Validate Code
# Payment processor connector
grace techspec adyen
grace techspec shopify --verbosegit clone <repository-url>
cd grace
uv sync --extra devuv run black src/
uv run mypy src/If you get Python version conflicts:
# Check Python version
python --version # Should be 3.9+
# Clear cache and reinstall
uv cache clean
uv syncIf LangGraph imports fail:
# Install core dependencies
uv add langgraph langchain langchain-core