git clone https://github.com/buildingopen/opengtm
cd opengtm
pip install -e ".[dev]"
cp .env.example .env
# Add your GEMINI_API_KEY to .env- Fork the repo and create a branch:
git checkout -b my-feature - Make your changes
- Test with real data using
GEMINI_API_KEYin your.env - Open a PR against
main
For significant changes (new modules, breaking changes), open an issue first to discuss.
- New ICP profiles — add to
ICP_PROFILESdict inqualify.py - New message frameworks — add patterns to
message.py, both EN and DE - CRM integrations — add adapters in
sync.py(HubSpot, Pipedrive, Airtable) - Language support — extend
_STRINGSdict inmessage.py - SERP volume — integrate Serper or DataForSEO in
keywords.pystage 6
- Python 3.10+
from __future__ import annotationsat top of every module- Use
DEFAULT_MODELfromopengtm/__init__.pyfor all Gemini model references, never hardcode model strings - All Gemini calls: try
google-genaiSDK first, fall back togoogle-generativeai - No external runtime dependencies beyond those in
pyproject.toml