跳转到主要内容
此内容尚未提供您的语言版本,正在以英文显示。

Deep Interview

技能 已验证 活跃

Socratic deep interview with mathematical ambiguity gating before explicit execution approval

目的

To transform vague ideas into crystal-clear specifications by systematically exposing assumptions and measuring clarity before any code is written or executed.

功能

  • Socratic questioning with mathematical ambiguity scoring
  • Brownfield/greenfield project analysis
  • Topological component enumeration and gating
  • Progressive disclosure of requirements
  • Structured specification generation

使用场景

  • Clarifying vague ideas before starting development
  • Ensuring precise understanding of requirements via Socratic dialogue
  • Validating clarity mathematically before committing to execution
  • Gathering detailed requirements for complex projects

非目标

  • Direct code execution or mutation
  • Brainstorming or option exploration (use omc-plan instead)
  • Quick fixes or single changes (delegate to executor or ralph)
  • Proceeding without explicit user approval for execution

工作流

  1. Parse user's idea and detect project type (brownfield/greenfield).
  2. Perform initial codebase exploration (brownfield) and context loading.
  3. Conduct Round 0 topology enumeration and confirmation.
  4. Iteratively ask targeted questions to reduce ambiguity, scoring clarity after each answer.
  5. Generate and crystallize the specification based on clarified requirements.
  6. Present execution options (refine with omc-plan, execute with autopilot, etc.) for user approval.

实践

  • Requirements Engineering
  • Specification Design
  • Interactive Refinement

先决条件

  • Claude Code
  • OpenAI API key or Claude Pro subscription

Documentation

  • info:Configuration & parameter referenceConfiguration options for ambiguity threshold and max rounds are documented, but specific details on how to configure them or precedence order are not explicitly detailed.

安装

请先添加 Marketplace

/plugin marketplace add Yeachan-Heo/oh-my-claudecode
/plugin install oh-my-claudecode@omc

质量评分

已验证
99 /100
1 day ago 分析

信任信号

最近提交1 day ago
星标33.7k
许可证MIT
状态
查看源代码