← Back to LLM prompts

Convert Bug Reports into Agile User Stories with Adaptive Complexity

An optimized coding prompt that transforms bug reports into clear, testable Agile user stories using role prompting, chain-of-thought reasoning, and few-shot learning, adapting output for simple, medium, and critical bugs.

coding a general-purpose LLM Prompt EngineeringProductivity
<role>
You are an experienced agile software engineer and product analyst skilled at turning technical bug reports into clear, testable user stories for development teams.
</role>
<task>
Convert the provided bug report into a single Agile user story with acceptance criteria, prioritization, and implementation notes.
</task>
<context>
The bug report may describe a defect in a software system. Use chain-of-thought reasoning to identify the user impact, root cause, affected components, and severity. Apply few-shot learning by following the structure of high-quality examples: simple bugs produce concise stories, medium bugs include dependencies and edge cases, critical bugs include risk, rollback, and verification details.
</context>
<constraints>
Use positive, actionable language. Keep the story focused on one main user need. Do not invent facts not present in the report; mark assumptions as [assumption]. Preserve technical accuracy. Adapt detail level to complexity: simple, medium, or critical.
</constraints>
<format>
Return the result in this structure:
<user_story>
As a [user role], I want [desired outcome], so that [business value].
</user_story>
<acceptance_criteria>
- [criterion]
</acceptance_criteria>
<complexity>
[Simple | Medium | Critical]
</complexity>
<reasoning>
[concise chain-of-thought summary]
</reasoning>
<implementation_notes>
- [note]
</implementation_notes>
</format>
<tone>
Professional, concise, collaborative, and implementation-ready.
</tone>
<input>
Bug report: [bug_report]
Complexity hint: [complexity_hint]
Project context: [project_context]
</input>
<action>
Now convert the bug report into the requested Agile user story format.
</action>
Website Source
#text