SF Hacks 2026 WinnerBest Hack for Climate Action

See the True Cost
of Every AI Query

Eco Mind is a suite of tools including a browser extension, MCP server, and VS Code extension that tracks real-time energy, water, CO₂, and token usage across every AI platform you use.

SEE IT IN ACTION

Measurable Impact

See the difference not just in dollars, but in computational efficiency.

Token Optimization

0kvs0k
Standard100%
Optimized-88%

Cost Analysis

$0.00vs$0.00
Standard100%
Optimized-83%

CO₂ Emissions

0gvs0g
Standard100%
Optimized-67%

Water Usage

0mlvs0ml
Standard100%
Optimized-47%

Real-Time Visibility

Watch your environmental impact update live as you interact with AI platforms

ECOMIND

EcoMind
FLASH

Input Tokens

0

0 saved

Output Tokens

0

0 saved

Energy

0.00

mWh

0.00 mWh saved

Water

0.00

mL

0.00 mL saved

Carbon

0.000

g

0.000 g saved

Energy Optimization

Semantic Cache

Zero tokens spent on repeated queries

Prompt Compression

Removes redundant input tokens

Concise Replies

Cuts output tokens by ~40%

// MISSION

The Invisible Scaling Problem

We are building the observability layer for the AI era, from browser extensions to MCP servers to VS Code sidebars.Every AI interaction has a cost. We make it visible.

10x

Energy Surge

An AI query consumes 10x more power than a standard search.

300M+

Tons CO2

Data centers now rival the aviation industry in emissions.

24/7

Resource Drain

Continuous model training demands millions of gallons of water.

// TEAM

Meet the Builders

Built in 24 hours at SF Hacks 2026 by a team passionate about making AI's environmental impact visible and actionable.

Nicolas Dos Santos

Nicolas Dos Santos

Software Engineer

UC San Diego

Mathematics-Computer Science student and SWE at UCSD. Serves in Cyber Defense Operations with the US Air Force National Guard.

Alfred Sjöqvist

Alfred Sjöqvist

ML & Cognitive Science

Stanford University

Pursuing an M.Sc in Computer Engineering with a B.Sc in Cognitive Science. Focusing on AI and Machine Learning at Stanford.

Matthew Garcia

Matthew Garcia

Full-Stack AI SWE

UC Riverside

Full-stack engineer specializing in scalable AI-driven applications with Python and Next.js. Career Accelerator Team Lead and hackathon enthusiast.

// HOW IT WORKS

Three steps to transparency

1

Install Eco Mind

Add the Chrome extension for browser AI, the MCP server for Claude Code, or the VS Code extension. Pick what fits your workflow.

2

Use AI as Usual

Chat with Gemini, ChatGPT, Claude, or code with Claude Code. Eco Mind works silently in the background.

3

See Your Footprint

Get real-time energy, water, CO₂, and token metrics, then optimize with AI-powered context reduction.

// FEATURES

Built for Awareness

Everything you need to understand and reduce the environmental footprint of your AI usage.

Browser Extension

A Chrome extension that tracks real-time energy, water, CO₂, and token usage as you chat on Gemini, Claude, and ChatGPT.

Live In-Page Widget

MCP Server

An MCP server for Claude Code that reads real session logs and generates Gemini-powered carbon footprint reports, with no manual tracking.

Claude Code Integration

VS Code Extension

A sidebar panel that tracks your Claude Code sessions right inside VS Code with real-time energy and carbon metrics.

Inline IDE Tracking

Context Optimization

Reduce token usage up to 90% with intelligent file summarization and signature extraction powered by Gemini.

Up To 90% Reduction

Multi-Platform

Works across Gemini, ChatGPT, Claude.ai, and Claude Code. One tool for your entire AI workflow.

Universal Compatibility

Privacy First

All tracking is done locally. We never store or transmit your prompts, responses, or code content.

Local Processing Only