Update app.py

app.py

@@ -6,6 +6,54 @@ import gradio as gr
 from pydantic import BaseModel, Field
 from typing import Optional, Literal
 
+echo_prompt_refiner = """You are an AI assistant implementing the ECHO (Self-Harmonized Chain of Thought) method to refine an initial prompt into a more relevant and comprehensive one. Given an initial prompt, follow these steps:
+
+1. Prompt Analysis:
+   - Analyze the initial prompt
+   - Identify key concepts, objectives, and constraints within the prompt
+   - Determine the prompt type (e.g., task-oriented, creative, analytical)
+
+2. Prompt Expansion:
+   - Generate 5-7 alternative versions of the prompt that explore different aspects or phrasings
+   - Ensure these versions cover various perspectives and potential interpretations
+   - Include a mix of more specific, more general, and differently focused versions
+
+3. Prompt Clustering:
+   - Group the generated prompts into 2-3 thematic clusters
+   - Identify the core themes or objectives represented by each cluster
+
+4. Demonstration Generation:
+   - For each cluster, outline how an AI might interpret and approach that prompt version
+   - Include potential reasoning steps and areas of focus, without generating actual responses
+
+5. Prompt Refinement:
+   - Review the demonstration outlines and identify strengths and weaknesses of each prompt version
+   - Refine each version, addressing potential misinterpretations and improving clarity
+   - Ensure each refined version maintains the original intent while enhancing specificity or broadening scope as needed
+
+6. Cross-Pollination:
+   - Identify effective elements from each refined version
+   - Integrate these elements to create a synthesized, improved prompt
+
+7. Final Prompt Synthesis:
+   - Combine the most effective elements from all refined versions
+   - Construct a final, comprehensive prompt that captures the essence of the original while incorporating improvements
+
+8. Output your results in a JSON format with the following keys:
+   - "initial_prompt": The original prompt provided
+   - "initial_prompt_evaluation": Your analysis of the initial prompt
+   - "expanded_prompts": An array of the alternative prompt versions you generated
+   - "prompt_clusters": An object containing the clustered prompt versions
+   - "demonstration_outlines": An object containing your approach outlines for each cluster, without actual responses
+   - "refined_prompts": An array of the refined prompt versions
+   - "refined_prompt": The final, synthesized prompt
+   - "explanation_of_refinements": An explanation of how the final prompt improves upon the initial one and any potential limitations or considerations for the refined prompt
+
+Ensure each step of your process is thorough and well-documented in the JSON output. Your final refined prompt should be clear, comprehensive, and effectively capture the intent of the initial prompt while addressing any identified shortcomings.
+
+Initial prompt: [Insert initial prompt here]
+
+Please provide your response in the specified JSON format."""
 
 metadone="""
 As an AI Prompt Enhancement Specialist, your mission is to elevate the given prompt using state-of-the-art prompting techniques while emphasizing the utilization of previously generated context. Analyze the input prompt and apply the following comprehensive approach to enhance it:
@@ -478,7 +526,7 @@ from typing import Optional, Literal
 
 class PromptInput(BaseModel):
     text: str = Field(..., description="The initial prompt text")
-    meta_prompt_choice: Literal["done","physics","morphosis", "verse", "phor","bolism"] = Field(..., description="Choice of meta prompt strategy")
+    meta_prompt_choice: Literal["star","done","physics","morphosis", "verse", "phor","bolism"] = Field(..., description="Choice of meta prompt strategy")
 
 class RefinementOutput(BaseModel):
     query_analysis: Optional[str] = None
@@ -502,6 +550,8 @@ class PromptRefiner:
             selected_meta_prompt = loic_metaprompt
         elif prompt_input.meta_prompt_choice == "done":
             selected_meta_prompt = metadone
+        elif prompt_input.meta_prompt_choice == "star":
+            selected_meta_prompt = echo_prompt_refiner
         else:
             selected_meta_prompt = advanced_meta_prompt
     
@@ -583,7 +633,7 @@ class GradioInterface:
             with gr.Row():
                 prompt_text = gr.Textbox(label="Type the prompt here")
             with gr.Row():
-                meta_prompt_choice = gr.Radio(["done","physics","morphosis", "verse", "phor","bolism"], label="Choose Meta Prompt", value="morphosis")
+                meta_prompt_choice = gr.Radio(["star","done","physics","morphosis", "verse", "phor","bolism"], label="Choose Meta Prompt", value="morphosis")
                 refine_button = gr.Button("Refine Prompt")
             with gr.Row():
                 gr.Markdown("### Initial prompt analysis")
@@ -623,7 +673,7 @@ class GradioInterface:
             gr.Examples(
                 examples=[
                     ["Tell me about that guy who invented the light bulb", "physics"],
-                    ["Explain the universe.", "done"],
+                    ["Explain the universe.", "star"],
                     ["What's the population of New York City and how tall is the Empire State Building and who was the first mayor?", "morphosis"],
                     ["List American presidents.", "verse"],
                     ["Write a story.", "bolism"],
@@ -631,7 +681,7 @@ class GradioInterface:
                     ["Is nuclear energy good?", "verse"],
                     ["How does a computer work?", "phor"],
                     ["How to make money fast?", "done"],
-                    ["how can you  prove IT0's lemma in stochastic calculus ?", "verse"],                    
+                    ["how can you  prove IT0's lemma in stochastic calculus ?", "star"],                    
                 ],
                 inputs=[prompt_text, meta_prompt_choice]
             )