cognitive-friction

Surface key risks and pause before high-stakes actions.

6|1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/Forexgod21/YVYC-Claude-Skills --skill cognitive-friction
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cognitive-friction
Source: https://github.com/Forexgod21/YVYC-Claude-Skills/tree/main/agentic/cognitive-friction
Command: npx skills add https://github.com/Forexgod21/YVYC-Claude-Skills --skill cognitive-friction

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Cognitive-friction teaches Claude to slow down at the right moments. Not everything deserves the same speed. A low-stakes draft gets executed fast. A high-stakes decision with real consequences deserves a pause — a moment where Claude surfaces what it is about to do, why, and what the user should consider before proceeding.

This approach ensures Claude acts deliberately when needed while remaining unobtrusive for routine tasks.

Core Features & Use Cases

  • Surface before proceeding: Claude explains what it's about to do and why.
  • One-question rule: Claude asks a single clarifying question when needed.
  • Risk naming: Claude identifies and names the specific risk.
  • Guardrails against alarm fatigue: friction is applied only when stakes deserve attention.

Use cases include high-stakes financial decisions, legal or regulatory reviews, security-sensitive actions, and scenarios with high uncertainty where deliberate thought adds value.

Quick Start

Install the SKILL.md into your Claude skills directory and begin activating friction on high-stakes tasks with the prescribed surface-and-pause flow.

Frequently Asked Questions about cognitive-friction

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add friction to high-stakes AI decisions to prevent irreversible errors?

To add friction to high-stakes AI decisions, you implement a surface-before-proceed rule that forces the model to explain its intended action, name specific risks, and pause for user confirmation before executing consequential or irreversible tasks.

What is cognitive friction in AI safety and risk management?

Cognitive friction in AI safety is a deliberate slowing mechanism that forces a model to surface key risks and clarify uncertainty before proceeding. It ensures deliberate decision-making on high-stakes tasks while remaining unobtrusive for routine operations.

How do I stop AI from rushing through legal, financial, or security-sensitive actions?

You stop AI from rushing by applying a progressive friction model with a one-question clarification rule. This requires the model to explicitly identify regulatory or security risks and ask a single clarifying question before taking action in highly uncertain scenarios.

Can I apply decision-making friction without causing alarm fatigue on routine tasks?

Yes, you can apply decision-making friction without alarm fatigue by implementing guardrails that trigger pauses only when stakes genuinely deserve attention. Low-stakes drafts execute fast, while irreversible strategic planning triggers a deliberate pause.

How do I integrate a reversibility gate and trust calibration into AI governance?

You integrate reversibility gates and trust calibration into AI governance by using optional companion skills alongside core friction rules. This combination validates action safety and adjusts trust levels before the model proceeds with high-stakes decisions.