Drip

Track and surface the estimated water cost of Claude interactions. Every query has a physical footprint - data centers need cooling, electricity needs generation. This skill adds environmental accountability by estimating water consumption per session and surfacing it at meaningful moments. Not to induce guilt, but to create awareness of the physical cost of intelligence.

62 stars 2 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add ElliotJLT/Claude-Skill-Potions --skill drip
Manual install
git clone --depth 1 https://github.com/ElliotJLT/Claude-Skill-Potions.git /tmp/Claude-Skill-Potions
mkdir -p ~/.claude/skills
cp -r /tmp/Claude-Skill-Potions/skills/drip ~/.claude/skills/drip

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Drip

<purpose>
AI feels weightless. Type a question, get an answer. But every token requires
compute, compute requires cooling, and cooling requires water. This skill
makes the invisible visible: the physical cost of conversation. Not to shame,
but to acknowledge that intelligence has a footprint.
</purpose>

## The Numbers (Honest Assessment)

<data-sources>
**What we know (2024-2025 research):**

Self-reported by providers (direct cooling only):
- Google Gemini: ~0.26ml per query
- OpenAI GPT-4: ~0.3ml per query

Academic estimates (including electricity generation water):
- 5-10ml per query for efficient models
- Up to 150ml for reasoning-heavy models like DeepSeek-R1

Per-token estimate (derived):
- ~0.5ml per 1,000 tokens (mid-range, including indirect)
- ~0.05ml per 1,000 tokens (direct cooling only)

**Why estimates vary:**
- Direct vs indirect water (cooling vs electricity generation)
- Regional data center efficiency (WUE ranges 0-3+ L/kWh)
- Model efficiency (1B vs 70B+ parameters)
- Query complexity (simple vs chain-of-thought reasoning)

**What we use:** 0.5ml per 1,000 tokens (conservative mid-range)
This includes indirect water from electricity but excludes hardware manufacturing.

Sources:
- UC Riverside/Colorado study (2023)
- "How Hungry is AI?" benchmark (2025)
- Google/OpenAI self-reported figures (2024-2025)
</data-sources>

## When To Surface

<triggers>
Surface water estimates at:

**Session milestones:**
Read the full skill on GitHub

Details

Skill name
drip
Path
skills/drip/SKILL.md
License
MIT
Repo updated
12 February 2026

FAQ

What does the Drip skill do?

Track and surface the estimated water cost of Claude interactions. Every query has a physical footprint - data centers need cooling, electricity needs generation. This skill adds environmental accountability by estimating water consumption per session and surfacing it at meaningful moments. Not to induce guilt, but to create awareness of the physical cost of intelligence.

How do I install the Drip skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add ElliotJLT/Claude-Skill-Potions --skill drip. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Drip popular?

It lives in ElliotJLT/Claude-Skill-Potions, which has 62 GitHub stars and 2 forks.

Related skills

Agent Audit

62

ElliotJLT/Claude-Skill-Potions

When reviewing, optimizing, or inheriting an LLM agent, systematically identify what's actually load-bearing vs. cruft. Instruments usage, ablates components, and strips the agent to minimum viable form. Answers: "What can we remove without degrading quality?" The goal is the leanest agent that still works.

AI, LLMs & Agents

Battle Plan

62

ElliotJLT/Claude-Skill-Potions

Before ANY significant development task (new feature, refactor, integration, migration), run a complete planning ritual by orchestrating other skills in sequence: rubber-duck (clarify scope) -> pre-mortem (assess risks) -> eta (estimate time) -> final confirmation. Do not start coding until the battle plan is approved.

Software Development

Breadcrumbs

62

ElliotJLT/Claude-Skill-Potions

Leaves notes for future Claude sessions in .claude/breadcrumbs.md. At session end or after significant work, records what was tried, what worked, what failed, and context for next time. On session start, reads breadcrumbs to understand history. Session-to-session memory without manual updates.

Productivity & Notes

Installing Skills System

43

obra/clank

Fork, clone to ~/.clank, run installer, edit CLAUDE.md

AI, LLMs & Agents26

Dispatching Parallel Agents

43

obra/clank

Use multiple Claude agents to investigate and fix independent problems concurrently

AI, LLMs & Agents26

Split Decision

62

ElliotJLT/Claude-Skill-Potions

When facing architectural decisions, technology choices, or strategic trade-offs, present options as a structured comparison and require explicit trade-off acknowledgment before proceeding. Triggers on words like "should we", "which approach", "what's the best way", or when Claude is about to recommend one approach over alternatives. Never present a single recommendation without showing viable alternatives first.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.