TouchGrass AI: An Open-Source AI Outdoor Mission Planner

Description and use

TouchGrass AI is an open-source outdoor mission planner that uses AI only to generate short missions. Users select time, activity, difficulty, and mode, receive a checklist, go outside, complete the mission, and return to mark it done. The stack includes React, FastAPI, SQLite, and an open-weight Gemma model with a deterministic fallback.

Problem

Digital products often keep users on screens; the goal was to create an AI that only serves as a brief planning tool and encourages users to spend time outdoors.

Outcome

The application successfully generates short outdoor missions, records completion, and updates user statistics, demonstrating that AI can facilitate reduced screen time.

Workflow

• Select available time (15, 30, or 60 minutes)
• Choose outdoor activity (Walking, Nature, Photography, Exercise, Social)
• Choose difficulty (Easy, Medium, Hard)
• Choose mode (Solo or With Friends)
• Backend creates structured mission
• App provides actionable checklist
• User goes outside and completes mission
• Mission is saved to history and statistics updated

Evidence

TouchGrass AI creates short, practical outdoor missions based on: Available time: 15, 30, or 60 minutes; Activity: Walking, Nature, Photography, Exercise, or Social; Difficulty: Easy, Medium, or Hard; Mode: Solo or With Friends

Source

The core interaction is intentionally simple: Choose → Generate → Go outside → Complete → Come back

Source

The project uses: Frontend: React + Vite; Backend: FastAPI + Python; Database: SQLite + SQLAlchemy; AI layer: Open-weight Gemma integration with a deterministic fallback

Source

For the deployed demo, I did not configure a hosted Gemma API key, so I have not claimed that the live deployment is currently performing real Gemma inference.

Source