paranoia

Enforce adversarial thinking to reveal hidden intent and manipulation in agents.

10|Updated Feb 26, 2026

Search Tags:#risk-assessment #threat-modeling #adversarial-thinking #systems-analysis #paranoia #mindstate

Ready to run

npx skills add https://github.com/olivierlesnicki/addhumanity --skill paranoia

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: paranoia
Source: https://github.com/olivierlesnicki/addhumanity/tree/main/skills/paranoia
Command: npx skills add https://github.com/olivierlesnicki/addhumanity --skill paranoia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reveal hidden intent and potential manipulation by enforcing adversarial thinking in agents.

Core Features & Use Cases

Paranoid mode surfaces gaps in reasoning by challenging conclusions.
Useful for threat modeling, risk assessment, and resilience testing in complex systems.
Use case: When you want an agent to scrutinize incentives and identify failure paths.

Quick Start

Ask the agent to map potential failure paths and hidden incentives in the current scenario.

Frequently Asked Questions about paranoia

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

FAQPage Schema

What is adversarial thinking in threat modeling and risk assessment?▼

Threat modeling with adversarial thinking works by enforcing paranoid reasoning to surface gaps in logic and challenge assumptions. It reveals hidden intent and potential manipulation by mapping potential failure paths across complex systems.

How do I map potential failure paths and hidden incentives in a system?▼

You map failure paths by asking an agent to apply adversarial thinking to scrutinize incentives and challenge conclusions in the current scenario. This surfaces gaps in reasoning to reveal hidden manipulation and failure paths.

Can I use adversarial thinking for resilience testing in complex systems?▼

Yes, adversarial thinking supports resilience testing in complex systems by enforcing paranoid reasoning that challenges conclusions and maps failure paths. It identifies hidden manipulation and scrutinizes incentives to test system robustness.

What's the best way to surface gaps in reasoning during risk assessment?▼

The best way to surface reasoning gaps during risk assessment is to enforce paranoid mode that actively challenges conclusions and maps failure paths. This adversarial thinking reveals hidden manipulation and scrutinizes incentives.

When should I not use paranoid mode for systems analysis?▼

Avoid paranoid mode for systems analysis when straightforward execution is needed without challenging conclusions. It enforces adversarial thinking to scrutinize incentives and map failure paths, which can complicate simple scenarios.