원클릭으로
bug-fix
Constrains the Agent to follow a structured bug fix process — locate → root cause → fix → regression verify.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Constrains the Agent to follow a structured bug fix process — locate → root cause → fix → regression verify.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
约束 Agent 必须遵循结构化的 Bug 修复流程:定位 → 根因分析 → 实施修复 → 回归测试验证。
约束 Agent 从正确性、安全性、性能和可维护性四个维度进行系统化的结构性代码审查。
约束 Agent 必须遵循结构化的功能特性开发流程:理解需求 → 技术设想 → 编写代码 → 自验自测。
约束 Agent 遵循标准流程生成结构化、高质量且要素齐全的产品需求文档。
约束 Agent 必须遵循安全的重构流程:在保证原有业务逻辑和功能完全等价的前提下,提升代码质量。
Constrains the Agent to generate structured Product Requirements Documents (PRDs) with quality and completeness guarantees.
| name | bug_fix |
| description | Constrains the Agent to follow a structured bug fix process — locate → root cause → fix → regression verify. |
| assets | {"checklist":"assets/checklist.yaml","template":"assets/template.yaml"} |
| references | {"guidelines":"references/guidelines.yaml","security":"references/security.yaml"} |
Constrains the Agent to follow the "locate → root cause analysis → fix → regression verification" structured process.
Before fixing a bug, you must check every item in assets/checklist.yaml.
Your output must strictly follow the format defined in assets/template.yaml.
Must include: root cause analysis, fix plan, fix code (diff format), impact analysis, regression checklist.
Vague suggestions like "try changing this" are not allowed.
If you need to execute commands to reproduce or verify a bug, check references/security.yaml first.
Do not execute any high-risk commands that could damage existing code or environment.
After fixing, audit your fix against references/guidelines.yaml.
Key checks: does the fix address the root cause? Does it introduce new bugs? Is the diff minimal?
Roll back and regenerate if non-compliant, up to 3 retries.