Carbon Footprint Tracker

A Flask web app that lets users log daily activities, calculates their carbon footprint, and rewards low-emission behavior with real tokens on the Algorand blockchain. Includes an AI-powered analyst (Groq API) that turns raw emissions data into plain-language feedback, and a peer-to-peer green-energy contract marketplace.

⚠️ Work in progress — core features run end-to-end, but this is not a polished/production build. See "Known limitations" below before judging it as a finished product.

What it does

• Daily emissions logging — users submit activity data (transport, energy use, etc.); the app computes net emissions and an eco-score.
• On-chain rewards — users earn "Green Credits" for low-emission days, paid out as real Algorand tokens to a wallet created for them on registration (algo/rewards.py, algo/ledger.py).
• On-chain ledger — each day's emission result is recorded as a note on an Algorand transaction, giving an auditable, tamper-resistant history (record_emission_on_chain).
• AI analyst — analyst/narrator.py generates a natural-language summary of a user's trends (best/worst day, vs. national/state average) using an LLM.
• Vendor marketplace — vendors can list renewable-energy contracts; consumers browse and buy them with Green Credits, with a Groq-generated recommendation on which contract fits their usage best.
• Leaderboard & streaks — gamifies consistent low-emission behavior.

Stack

• Backend: Python, Flask
• Blockchain: Algorand (Python SDK) — wallet funding, token transfers, on-chain ledger notes
• AI: Groq API (Llama 3.1) for narrative generation and marketplace recommendations
• Storage: local JSON-based storage (see storage/, data/)
• Frontend: Flask templates (templates/)

Setup

pip install -r requirements.txt
export GROQ_API_KEY=your_key_here
python app.py

Runs two dev servers on ports 3000 and 5000.

Known limitations

• Startup performs a remote authentication check against a file in a separate personal repo; this needs to be removed or made optional so the app runs standalone for anyone cloning it.
• Secret key and other config values are hardcoded — move to environment variables before any real deployment.
• No automated tests yet.
• UI is functional, not polished.

Status

Actively being cleaned up — algorithm, consensus, and analyst modules are functional; hardening (secrets, auth gate, tests) is in progress.