Everything you need to install, configure, and run Loopmark, plus how to contribute back.
Requires Python 3.10+ and an OpenAI API key.
git clone https://github.com/loopmark-opensource/loopmark-agent.gitcd loopmark-agentpython3 -m venv .venvsource .venv/bin/activatepip install -r requirements.txtcp .env.example .envThen set OPENAI_API_KEY in .env and run:
python3 main.pyOnly OPENAI_API_KEY is required. Everything else is optional.
python3 main.pyStart an interactive chatpython3 main.py --verboseShow which sub-agent handled each requestpython3 main.py statsView a summary of stored datapython3 scheduler.pyRun the auto-posting schedulerloopmark-agent/├── main.py# Interactive CLI├── scheduler.py# Auto-posting daemon├── agents/│ ├── graph.py# Router + orchestrator│ └── sub_agents.py# Complaints, Posting, Funnel├── tools/│ ├── complaint_tools.py│ ├── social_tools.py│ ├── email_tools.py│ ├── publisher_tools.py│ └── funnel_tools.py└── data/# Auto-created JSON storageLoopmark is MIT licensed and open to contributions. We use a simple feature-branch workflow.
git checkout -b feature/my-change# make changes, commit, pushgit push origin feature/my-changeCreate your own copy on GitHub.
git checkout -b feature/my-change
Keep changes focused and match the existing code style.
Describe your change clearly against main.
Copy .env.example to .env and add your real API key.
Twitter write access requires a paid API tier. Use Buffer instead.
Make sure the scheduler is running and scheduled_date is today or earlier.