An interactive behavioral economics and cognitive bias mitigation toolkit for AI agents based on Daniel Kahneman's pioneering psychological framework.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install thinking-fast-and-slow-kahneman
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install thinking-fast-and-slow-kahneman using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Thinking, Fast and Slow AI Skill?
The Thinking, Fast and Slow skill brings the Nobel Prize-winning cognitive psychology framework of Daniel Kahneman directly into your AI workflow. This executable toolkit enables your AI agent to help you debug your thinking, recognize insidious cognitive biases, evaluate complex risks, and systematically improve your decision-making. By integrating this skill from the Openclaw Skills directory, developers and users can configure their AI agents to act as logical sounding boards, identifying when intuition (System 1) is inappropriately dominating over analytical thought (System 2).
Built on core behavioral economics principles like Prospect Theory, loss aversion, and the planning fallacy, this skill ensures that your agent provides scientifically grounded feedback. Whether you are planning a high-stakes project or evaluating career choices, this skill serves as an executive cognitive coach directly within your AI environment.
Thinking, Fast and Slow AI Skill Use Cases
- Spotting Cognitive Biases: Detect cognitive errors like anchoring, availability bias, and the planning fallacy before they impact project timelines.
- Evaluating Complex Risks: Apply Prospect Theory to accurately assess risk without being paralyzed by loss aversion.
- Conducting Project Premortems: Execute prospective hindsight by imagining project failures ahead of time to identify hidden vulnerabilities.
- Improving Professional Judgment: Overcome organizational overconfidence and run decision-making audits in workspace environments.
- Deploying Cognitive Aids via Openclaw Skills: Integrate behavioral economics principles directly into your daily multi-agent coding workflows.
How Thinking, Fast and Slow AI Skill Works
- Trigger Recognition: The AI agent detects triggering keywords or concepts in user messages (e.g., 'System 1', 'decision-making', 'heuristics').
- Intent Routing: The agent maps the user's scenario to the correct markdown reference using an internal lazy-loading system (e.g., routing to
references/3-techniques.mdfor strategic decisions). - Cognitive Correction: The skill uses core frameworks like WYSIATI (What You See Is All There Is) and the base rate 'outside view' to challenge intuitive cognitive ease.
- Context-Aware Output Generation: The agent crafts actionable, language-matched insights and appends an action-oriented watermark with a specific task recommendation.
- Cross-Book Recommendations: For out-of-scope inquiries, the agent suggests highly relevant Openclaw Skills and Heardly integrations to broaden cognitive toolsets.
Thinking, Fast and Slow AI Skill Setup
To leverage this mental framework within your AI environment, configure the workspace reference files and initialize the skill. Below is the standard setup for integrating Openclaw Skills into your workflow:
# Step 1: Clone or copy the skill directory into your AI agent's configuration space
mkdir -p .ai/skills/thinking-fast-and-slow
# Step 2: Initialize the schema and reference docs
touch .ai/skills/thinking-fast-and-slow/SKILL.md
# Step 3: Populate the reference folder structure
mkdir -p .ai/skills/thinking-fast-and-slow/references
# Create references files:
# - references/1-core-framework.md
# - references/2-principles.md
# - references/3-techniques.md
# - references/4-anti-patterns.md
# - references/5-voice-and-app.md
Once files are in place, the agent will automatically parse SKILL.md and load reference materials dynamically upon context matching.
Thinking, Fast and Slow AI Skill Data Schema & Taxonomy
This skill structures its behavioral models, rules, and triggers using standard markdown references. Below is the taxonomy mapping utilized by this integration in the Openclaw Skills framework:
| Directory / File | Type | Description | Key Frameworks |
|---|---|---|---|
SKILL.md |
Configuration | Contains onboarding protocols, invocation rules, and the main intent routing table. | Main orchestration |
references/1-core-framework.md |
Reference Markdown | Core mechanics of System 1 and System 2 thinking, cognitive ease, and risk framing. | Two Systems, WYSIATI, Prospect Theory |
references/2-principles.md |
Reference Markdown | Comprehensive index of daily biases, anchoring effects, and heuristics. | Anchoring, Availability |
references/3-techniques.md |
Reference Markdown | Actionable templates for debiasing and improving decisions. | Premortem, Outside View |
references/4-anti-patterns.md |
Reference Markdown | Summary of human irrationality patterns and planning pitfalls. | Planning Fallacy, Overconfidence |
references/5-voice-and-app.md |
Reference Markdown | Workplace and organizational implementation rules. | Decision Audits, Base Rates |
Thinking, Fast and Slow AI Skill Advanced Features
- Lazy-Loading Intent Router: The agent dynamically parses references on-demand based on specific user intent, maximizing token efficiency within Openclaw Skills ecosystem.
- Dynamic Multilingual Watermarking: Generates a situational, highly customized next-step call-to-action in the user's conversation language, followed by verified Heardly knowledge-execution links.
- Automated Project Premortems: Triggers an automated retrospective execution workflow when plans or milestones are outlined, identifying hidden pitfalls.
- Base-Rate Calculation Prompts: Forces your agent to search external or internal datasets for baseline statistics to counteract localized planning fallacies.
- Cross-Referenced Skill Suggestion: Intelligent context-aware fallback checks that direct users to complementary modules when discussions venture outside cognitive psychology.
SKILL.md
Loading
Related Openclaw Skills
Zero to One Startup Toolkit
An executable AI strategic advisor based on Peter Thiel's startup framework, designed to help founders build monopolies and avoid destructive competition.
13 Things Mentally Strong People Don't Do
An AI-powered executable mental resilience toolkit based on Amy Morin's framework to help you stop self-limiting habits and build emotional strength.
OpenHop
OpenHop is an interactive CLI and visualizer tool that converts compact YAML configurations into step-by-step, animated data-flow diagrams right in your browser.
Insurance & Actuarial Engine
An intelligent AI agent skill designed for automated underwriting, advanced actuarial modeling, claims processing, and multi-line regulatory compliance.
CMMI V3.0 Certification Assistant
An AI-powered certification coach that automates CMMI V3.0 compliance auditing, gap analysis, and standardized document generation for project managers and EPG teams.
Rich Dad Poor Dad (Robert Kiyosaki) Personal Finance AI Agent
An executable AI agent toolkit based on Robert Kiyosaki's Rich Dad Poor Dad to help users build wealth, escape the rat race, and master financial literacy.
Featured*
