Thinking Tools for AI Agents¶

The Thinking Tools provide structured cognitive frameworks for systematic problem analysis and decision-making in AI agents.

🤖 AI Agent Setup¶

from enhancedtoolkits import ThinkingTools

# Initialize for your AI agent
thinking = ThinkingTools(
    enable_bias_detection=True,      # Detect cognitive biases
    enable_quality_assessment=True,  # Assess thinking quality
    thinking_depth=3                 # Analysis depth level
)

# Register with your agent
agent.register_tools([thinking])

⚙️ Configuration Options¶

🧠 Available Functions¶

Your AI agent will have access to these functions:

think()¶

Process thoughts using structured cognitive frameworks.

Parameters: - agent_or_team: Agent instance for session tracking - thought: The thought or problem to analyze - thinking_type: Type of thinking framework to apply - context: Additional context for analysis

Thinking Types: - "analysis" - Systematic breakdown and examination - "synthesis" - Combining elements into coherent whole - "evaluation" - Critical assessment and judgment - "reflection" - Self-examination and metacognition - "planning" - Strategic planning and goal setting - "problem_solving" - Systematic problem resolution - "creative" - Creative and innovative thinking - "critical" - Critical analysis and reasoning

analyze_thinking_quality()¶

Assess the quality of thinking processes.

detect_thinking_biases()¶

Identify cognitive biases in thinking patterns.

🎯 AI Agent Integration Examples¶

OpenAI Function Calling¶

import openai
from enhancedtoolkits import ThinkingTools

thinking = ThinkingTools()

# Get function schema for OpenAI
tools = [thinking.get_openai_schema()]

response = openai.ChatCompletion.create(
    model="gpt-4",
    messages=[{
        "role": "user", 
        "content": "Analyze the pros and cons of remote work"
    }],
    tools=tools,
    tool_choice="auto"
)

Agno Framework¶

from agno.agent import Agent
from enhancedtoolkits import ThinkingTools

agent = Agent(
    name="Strategic Analyst",
    model="gpt-4",
    tools=[ThinkingTools(thinking_depth=5)]
)

# Agent can now use thinking functions
response = agent.run("Think through the strategic implications of AI adoption")

🔧 Production Configuration¶

Basic Setup¶

thinking = ThinkingTools()

Advanced Setup¶

thinking = ThinkingTools(
    enable_bias_detection=True,
    enable_quality_assessment=True,
    thinking_depth=5,
    instructions="Focus on strategic and long-term implications"
)

Environment Variables¶

# Optional: Set logging level
LOG_LEVEL=INFO

🛡️ Features¶

• 8 Thinking Types: Analysis, synthesis, evaluation, reflection, planning, problem-solving, creative, critical
• Bias Detection: Automatic identification of cognitive biases
• Quality Assessment: Depth, clarity, and evidence integration analysis
• Session Tracking: Maintains thinking patterns and progression
• Structured Output: Consistent, analyzable thinking results

🔍 Example Agent Interactions¶

Agent Query: "Think through the decision to expand into international markets"

Thinking Tool Response:

{
  "thinking_type": "analysis",
  "analysis": {
    "factors": ["Market size", "Competition", "Regulatory environment"],
    "considerations": ["Cultural differences", "Currency risks", "Local partnerships"],
    "framework": "Systematic market analysis"
  },
  "quality_assessment": {
    "depth": "High",
    "clarity": "Clear",
    "evidence_integration": "Good"
  },
  "detected_biases": [],
  "recommendations": ["Conduct market research", "Assess regulatory requirements"]
}

📊 Monitoring¶

Enable detailed logging to monitor thinking processes:

import logging
logging.basicConfig(level=logging.DEBUG)

thinking = ThinkingTools(debug=True)

🚀 Next Steps¶

1. Initialize ThinkingTools with your preferred configuration
2. Register with your AI agent framework
3. Test with sample thinking queries
4. Monitor thinking quality and bias detection
5. Adjust thinking depth and bias detection as needed

The Thinking Tools help your AI agent develop more structured, unbiased, and high-quality thought processes for complex decision-making scenarios.