Touch Grass Sports AI – Offline AI That Gets You Outside

Description and use

Touch Grass Sports AI – a fully offline solution that generates personalized outdoor fitness challenges without sending data to the cloud.

Problem

Users want personalized outdoor sports challenges that keep them off screens and in real life, but existing solutions rely on cloud APIs and data transmission.

Outcome

The solution delivers fully offline, privacy‑preserving, customizable outdoor fitness challenges, enabling users to stay active without internet or data leakage.

Workflow

• 1. User selects fitness level, sport, available time, and location via Gradio UI or CLI.
• 2. The system chooses either a local LLM (Ollama) or a rule‑based engine based on availability.
• 3. Strict system prompts ensure the model outputs only outdoor activities that include a touch‑grass action.
• 4. The challenge is presented to the user for execution.
• 5. Optionally, users can swap models, edit prompts, or extend the challenge library locally.

Evidence

Touch Grass Sports AI is a fully offline open-source AI that generates personalized outdoor sports challenges.

Source

Works in two modes: Local LLM (Ollama) for creative challenges; Pure rule-based mode (no model needed).

Source

Runs completely offline (useful on trails and parks with no signal).

Source