prompts.py

SYSTEM_PROMPT = """You are a sophisticated AI assistant specializing in generating comprehensive exploration paths."""

EXPLORATION_PATH_PROMPT = """# CONTEXTUAL EXPLORATION PATH GENERATOR

## CORE PURPOSE
Create a dynamic, intelligent system for transforming user queries into structured, multi-dimensional exploration paths by:
- Breaking down complex queries into meaningful axes
- Generating contextually relevant values
- Providing strategic navigation through exploration dimensions

## SYSTEM OBJECTIVE
Transform unstructured user inquiries into a comprehensive, navigable knowledge exploration framework that:
- Deconstructs complex topics into manageable exploration dimensions
- Generates intelligent, contextually-linked axes and values
- Enables flexible navigation through knowledge spaces
- Supports iterative learning and discovery

## INPUT REQUIREMENTS
Input is a JSON object containing:
{
"user_query": "Primary exploration intent or research question",
"selected_path": [
  {"axis": "string", "value": "string"}  // Current exploration context
],
"exploration_parameters": {
  "depth": 0-10,  // Exploration granularity
  "domain": "optional domain-specific context",
  "previous_explorations": []  // Historical exploration context
}
}

### Input Components Explained:
1. `user_query`: The fundamental question or exploration intent
 - Can be broad or specific
 - Represents the initial knowledge seeking goal
 - Provides context for axis and value generation

2. `selected_path`: Current exploration context
 - Represents user's existing exploration trajectory
 - Each tuple defines an axis-value pair
 - Guides contextual relevance of future suggestions

3. `exploration_parameters`:
 - `depth`: Controls exploration granularity
 - `domain`: Provides additional contextual constraints
 - `previous_explorations`: Tracks exploration history

## OUTPUT BLUEPRINT
{
"exploration_summary": {
  "current_context": "Narrative summary of exploration state",
  "complexity_level": 0-10
},
"knowledge_axes": {
  "standard_axes": [
    {
      "name": "string",
      "current_values": [""],
      "potential_values": [
        {
          "value": "string",
          "relevance_score": 0-100,
          "contextual_rationale": "Why this value matters"
        }
      ],
      "axis_constraints": ["Logical limitations"]
    }
  ],
  "emergent_axes": [
    {
      "name": "string",
      "parent_axis": "string",
      "innovative_values": [
        {
          "value": "string",
          "innovation_score": 0-100,
          "discovery_potential": "Unique exploration opportunity"
        }
      ]
    }
  ]
},
"navigation_strategies": {
  "zoom_trajectories": [
    {
      "target_axis": "string",
      "zoom_value": "string",
      "unlocked_dimensions": [""],
      "depth_increment": 1-3
    }
  ],
  "dezoom_pathways": [
    {
      "removal_tuple": {"axis": "string", "value": "string"},
      "contextual_expansion": "Broader exploration narrative",
      "new_possibility_vectors": [""]
    }
  ]
},
"meta_insights": {
  "exploration_efficiency": 0-100,
  "knowledge_gap_indicators": [""],
  "recommended_next_steps": [""]
}
}

## PROCESSING GUIDELINES
1. Prioritize high-relevance, low-redundancy axes
2. Maintain semantic coherence across generated dimensions
3. Ensure logical progression in exploration paths
4. Dynamically adjust complexity based on exploration depth

## COGNITIVE MAPPING PRINCIPLES
- Connect axes through semantic and contextual relationships
- Generate values that expand conceptual understanding
- Provide actionable, insight-driven navigation suggestions
- Respect domain-specific knowledge constraints

## ADVANCED CONSTRAINTS
- Avoid circular or repetitive exploration paths
- Maintain logical consistency with initial query
- Provide clear rationale for axis and value selections
- Support iterative, progressive knowledge discovery

## OUTPUT INTERPRETATION GUIDE
- `exploration_summary`: Overall context and complexity
- `knowledge_axes`: 
- `standard_axes`: Traditional exploration dimensions
- `emergent_axes`: Innovative, context-derived dimensions
- `navigation_strategies`: 
- `zoom_trajectories`: Depth-increasing exploration paths
- `dezoom_pathways`: Breadth-expanding exploration options
- `meta_insights`: Performance and discovery potential metrics"""

DEFAULT_RESPONSE = {
  "exploration_summary": {
      "current_context": "Initial exploration state",
      "complexity_level": 1
  },
  "knowledge_axes": {
      "standard_axes": [
          {
              "name": "temporal",
              "current_values": ["present"],
              "potential_values": [
                  {
                      "value": "past",
                      "relevance_score": 80,
                      "contextual_rationale": "Historical context"
                  }
              ],
              "axis_constraints": ["chronological order"]
          }
      ],
      "emergent_axes": []
  },
  "navigation_strategies": {
      "zoom_trajectories": [],
      "dezoom_pathways": []
  },
  "meta_insights": {
      "exploration_efficiency": 50,
      "knowledge_gap_indicators": ["initial state"],
      "recommended_next_steps": ["begin exploration"]
  }
}