Thomaszhou22

Agent Carbon Tracker — AI skill for Claude Code

AI community

Track carbon footprint of AI agent operations.

How to install Agent Carbon Tracker

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Thomaszhou22/agent-carbon-tracker and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agent Carbon Tracker does

Track carbon footprint of AI agent operations. EcoLogits-based LLM energy estimation for ESG reporting. Works with OpenClaw, Claude Code, Cursor, and any agent platform.

Alternatives in AI

Manifest — Real-time cost observability for OpenClaw agents — track tokens, costs, messages, and model usage 4.1k ★
Golutra — Multi-agent AI orchestration platform for automation, workflows, and developer tools 3.8k ★
TokenTracker — Local-first AI token usage & cost tracker for 31 coding tools incl 1.4k ★

README

Agent Carbon Footprint Tracker

[English](#english) | [中文](./README_CN.md)

[![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE) [![GitHub stars](https://img.shields.io/github/stars/Thomaszhou22/agent-carbon-tracker?style=social)](https://github.com/Thomaszhou22/agent-carbon-tracker/stargazers) [![Methodology](https://img.shields.io/badge/methodology-EcoLogits-blue.svg)](https://ecologits.ai) [![OpenClaw](https://img.shields.io/badge/OpenClaw-compatible-blue)](https://openclaw.ai) [![Claude Code](https://img.shields.io/badge/Claude%20Code-compatible-purple)](https://claude.ai) [![Cursor](https://img.shields.io/badge/Cursor-compatible-orange)](https://cursor.sh) [![Platform](https://img.shields.io/badge/platform-all%20agent%20platforms-green)]

Introduction

Every LLM call, every shell command, every browser action an AI agent performs has an energy cost. As agent usage scales to millions of daily operations, these costs add up. Enterprises need ESG reports. Developers need to know which models are cheapest to run. Users deserve transparency.

Agent Carbon Footprint Tracker estimates energy consumption and CO2 emissions for every operation an AI agent performs, using the EcoLogits bottom-up Life Cycle Assessment methodology (ISO 14044). No hardware access needed, no external API keys, no cloud telemetry. Just open-source math.

Features

LLM call estimation — Calculates GPU energy from model active parameters and token counts, using published EcoLogits regression models
Exec command estimation — CPU TDP × utilization × duration for local shell commands
Browser operation estimation — Page-complexity-weighted power draw for browser automation
30+ model database — Pre-configured parameters for GPT-4/4o/o1/o3, Claude 3/3.5/4, Gemini, DeepSeek, Llama, Mistral, Qwen, GLM, and more
Daily / weekly / monthly reports — Formatted markdown or JSON output with trend