doubt

Force re-evaluation of evidence and alternative interpretations before asserting claims.

10|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/olivierlesnicki/addhumanity --skill doubt-olivierlesnicki
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: doubt
Source: https://github.com/olivierlesnicki/addhumanity/tree/main/skills/doubt
Command: npx skills add https://github.com/olivierlesnicki/addhumanity --skill doubt-olivierlesnicki

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Doubt helps agents avoid premature certainty by challenging initial conclusions, forcing explicit justification and search for alternative explanations.

Core Features & Use Cases

  • Encourages re-evaluation of assumptions and explicit evidence checks when faced with high-stakes decisions.
  • Applicable in debugging reasoning, risk assessment, and strategy planning where flawed certainty is risky.
  • Useful when you want to slow-roll output and surface hidden assumptions before committing to a claim.

Quick Start

Use doubt to challenge the first explanation and surface alternative interpretations before deciding.

Frequently Asked Questions about doubt

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

FAQPage Schema
How do I reduce overconfidence in AI reasoning and prevent premature conclusions?

Reduce overconfidence by forcing the agent to question its initial conclusions, requiring explicit evidence checks and tracking alternative interpretations before asserting a claim. This prevents premature certainty in reasoning tasks.

What is the best way to challenge initial assumptions during critical analysis and hypothesis generation?

Challenge initial assumptions by applying explicit re-evaluation of evidence and searching for alternative explanations during critical analysis. This surfaces hidden assumptions and flags unresolved questions before committing to a conclusion.

How do I surface hidden assumptions in strategy planning and risk assessment?

Surface hidden assumptions in strategy planning and risk assessment by enforcing re-evaluation of evidence and tracking alternative interpretations before committing to a claim. This slows down output to interrogate conclusions.

Does this approach work for high-stakes decisions where flawed certainty is risky?

Yes, this approach works for high-stakes decisions by encouraging re-evaluation of assumptions and explicit evidence checks. It mitigates the risk of flawed certainty in tasks like risk assessment and strategy planning.

When should I not use critical thinking prompts that slow-roll agent output?

Avoid using critical thinking prompts that slow-roll output when immediate, direct responses are required without the need to interrogate conclusions or track alternative interpretations, as the enforced re-evaluation adds processing overhead.

Can I use doubt to debug reasoning errors in hypothesis generation tasks?

Yes, you can use doubt to debug reasoning errors by challenging the first explanation and surfacing alternative interpretations. It forces the agent to justify its claims and flag unresolved questions during hypothesis generation.