Open-source AI and citizen science for secondary schools.
River AI Rangers turns school science and geography lessons into live environmental investigations. Students work with real water-quality data from a local river, use an AI assistant to interpret what the numbers mean for ecosystems and human health, and produce genuine civic outputs — letters to councillors, community briefings, school assemblies — directed at real audiences.
Built for secondary school teachers and environmental educators, so that any school near a river, a stream, or a wetland can turn water-quality data into meaningful science, genuine literacy, and real community action.
No installation. No subscriptions. Free to use.
- 📋 Full lesson plans (KS3/KS4 / ages 11–16)
- 📝 Printable pupil worksheets
- 💬 AI prompt card library for educators
- 📊 Sample river water-quality data (CSV)
- 🌊 Live AI assistant — river-ai-rangers.streamlit.app
- Secondary school science and geography teachers (KS3/KS4)
- Environmental educators and river or ocean action groups
- Developers interested in open-source AI for education
- Anyone who believes water deserves better
- Secondary-school first — lessons designed for ages 11–16, KS3/KS4 curriculum-aligned
- Action-oriented — every lesson ends with a real civic output for a real audience
- Critical AI literacy — students are explicitly taught to question and evaluate AI responses
- Open-source only — MIT licence, free to adapt and redeploy
- Place-based — anchored in the specific river nearest the school
- The teacher uploads a CSV from the local river monitoring group
- The AI assistant reads the latest readings and is ready to answer questions in context
- Students investigate the data, question the AI, and form their own conclusions
- Every session ends with a civic output: a letter, a poster, a briefing, or a speech
👉 river-ai-rangers.streamlit.app
Powered by Anthropic Claude · Runs in any browser · No installation required
To run your own instance, add your ANTHROPIC_API_KEY to Streamlit Cloud App Settings → Secrets.
Student questions and river data are processed by Anthropic's Claude API to generate AI responses. No personally identifiable information should be entered into the chat. River water-quality data is processed transiently to answer each question and is not stored by this application beyond the user's browser session.
Schools deploying this app for students should satisfy themselves that its use is consistent with their own data processing policies. For data processing details, see Anthropic's privacy policy.
| File | Description |
|---|---|
app.py |
Streamlit AI assistant (main app) |
requirements.txt |
Python dependencies |
sample_river_data.csv |
Example CSV from a river monitoring group |
lesson-plan-01.md |
Session 1 — Meet the River (60 min) |
worksheet-01.md |
Pupil worksheet for Session 1 |
ai-prompt-library-01.md |
Prompt card sets A–E |
River AI Rangers was founded by Noelia Rufete Gil, researcher in AI, citizen science and environmental education.
The programme was submitted to the Convene4Climate Global Awards 2026 in the Technology for Climate category.
MIT licence — adapt, translate, and redeploy freely.