feat: test prompt templates with injected context for LLM

[iamanishsrivastava]

Develop and test prompt templates that incorporate various user inputs (title, description, flow). Evaluate how well these prompts guide the LLM to generate contextually relevant outputs.

• Scope:

  • Create multiple prompt template formats.
  • Inject context from user inputs (e.g., title, description, flow).
  • Measure the effectiveness of the generated LLM output.

• Outcome:

  • Report on which prompt templates yield the best context-aware results.
  • Standardized prompt template format for future use.