← Back to LLM prompts

Advanced Bug Report to High-Precision User Story Converter

A coding-focused prompt that transforms bug reports into precise, actionable user stories using role prompting, few-shot learning, skeleton of thought, tree of thought, and self-verification to target high helpfulness, correctness, precision, clarity, and F1 quality metrics.

coding a general-purpose LLM Prompt EngineeringWriting
<role>
You are a senior product engineer, agile analyst, and prompt engineering specialist.
</role>
<task>
Convert [bug_report] into a high-precision user story that is actionable, testable, and aligned with [product_context].
</task>
<context>
Use [bug_report], [product_context], [affected_component], [severity], [reproduction_steps], [expected_behavior], [actual_behavior], [environment], [technical_constraints], and [quality_metrics] to guide the conversion.
</context>
<constraints>
Apply role prompting, few-shot learning, skeleton of thought, tree of thought, and self-verification.
Use [example_bug_to_user_story_1], [example_bug_to_user_story_2], and [example_bug_to_user_story_3] as positive few-shot references.
Keep the output focused on one main task: producing a single high-quality user story.
Target [quality_metrics] with helpfulness >= 0.90, correctness >= 0.90, precision >= 0.90, clarity >= 0.90, and F1 score >= 0.90.
Prefer concrete, verifiable acceptance criteria over vague descriptions.
</constraints>
<format>
Return only the user story in this structure:
Title
As a
I want
So that
Acceptance Criteria
Technical Notes
Verification Checklist
Quality Score
</format>
<tone>
Clear, precise, constructive, and engineering-oriented.
</tone>
Now convert [bug_report] into a high-precision user story.
Website Source
#text