src.llm_interpreter.llm.prompts

Prompt templates for AMMM LLM report generation.

This module implements structured-first prompting: 1. First extract structured insights from evidence 2. Then generate narrative from structured data

Author: AMMM Team Created: 2025-04-10 Last Modified: 2025-04-10

Module Contents

src.llm_interpreter.llm.prompts.format_evidence(insights_dict: dict[str, Any]) → str

Format insights dictionary as readable evidence for LLM.

Parameters:

insights_dict – Dictionary containing structured insights

Returns:

Formatted evidence string

src.llm_interpreter.llm.prompts.create_extraction_prompt(prompt_template: str, evidence: dict[str, Any]) → str

Create an extraction prompt with formatted evidence.

Parameters:

• prompt_template – Template string with {evidence} placeholder

• evidence – Dictionary of evidence data

Returns:

Formatted prompt

src.llm_interpreter.llm.prompts.create_narrative_prompt(prompt_template: str, structured_data: dict[str, Any]) → str

Create a narrative generation prompt with structured data.

Parameters:

• prompt_template – Template string with {structured_data} placeholder

• structured_data – Dictionary of structured insights

Returns:

Formatted prompt