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 Dive

技能 已验证 活跃

2-stage pipeline: trace (causal investigation) -> deep-interview (requirements crystallization) with 3-point injection

目的

To systematically uncover the root cause of a problem and then precisely define the necessary requirements, bridging the gap between investigation and actionable plans.

功能

  • 2-stage pipeline: trace (causal investigation) -> deep-interview (requirements crystallization)
  • 3 parallel trace lanes for comprehensive investigation
  • 3-point injection mechanism to enrich interview context
  • State persistence for resume resilience
  • Seamless handoff to planning and execution skills

使用场景

  • Investigating unknown root causes of bugs or system failures
  • Clarifying requirements for complex features before development
  • Understanding existing system behavior before making changes
  • When users request a 'deep dive' or 'trace and interview'

非目标

  • Directly executing code or making changes
  • Replacing standalone trace or deep-interview skills for specific use cases
  • Acting as a replacement for a full PRD or spec when one already exists

工作流

  1. Parse user input and detect project type (brownfield/greenfield).
  2. Generate 3 trace lane hypotheses.
  3. Present hypotheses for user confirmation.
  4. Execute trace lanes in parallel (or sequentially) with team mode.
  5. Synthesize trace findings, identify critical unknowns, and save trace results.
  6. Initiate deep-interview phase with injected trace context and initial questions.
  7. Conduct iterative, Socratic interview until ambiguity threshold is met.
  8. Generate final spec document with trace findings.
  9. Present execution options (ralplan, autopilot, etc.) via workflow pre-flight checks.

实践

  • Causal investigation
  • Requirements gathering
  • Workflow orchestration
  • System analysis

先决条件

  • Claude Code CLI
  • Claude Max/Pro subscription or Anthropic API key
  • tmux (for certain features like rate limit waiting)

Documentation

  • info:Configuration & parameter referenceWhile the SKILL.md describes workflow and state persistence, explicit documentation for all potential configuration parameters, defaults, and precedence orders is not 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

类似扩展

Ralplan

99

Consensus planning entrypoint that auto-gates vague ralph/autopilot/team requests before execution

技能
Yeachan-Heo

Ce Brainstorm

100

Explore requirements and approaches through collaborative dialogue, then write a right-sized requirements document. Use when the user says "let's brainstorm", "what should we build", or "help me think through X", presents a vague or ambitious feature request, or seems unsure about scope or direction -- even without explicitly asking to brainstorm.

技能
EveryInc

Cli Creator

100

Build a composable CLI for Codex from API docs, an OpenAPI spec, existing curl examples, an SDK, a web app, an admin tool, or a local script. Use when the user wants Codex to create a command-line tool that can run from any repo, expose composable read/write commands, return stable JSON, manage auth, and pair with a companion skill.

技能
openai

Context Mode Ops

100

使用并行子代理军队管理 context-mode GitHub 问题、PR、发布和营销。为每个任务编排 10-20 个动态代理。在分类问题、审查 PR、发布版本、撰写 LinkedIn 帖子、宣布发布、修复错误、合并贡献、验证 ENV 变量、测试适配器或同步分支时使用。

技能
mksglu

Fixflow

100

使用严格的交付工作流执行编码任务:构建完整计划、分步实现、持续运行测试,并默认在每一步 (`per_step`) 后提交。当用户要求行为驱动交付或需求不明确时,支持显式提交策略覆盖 (`final_only`, `milestone`) 和可选的 BDD(给定/当/则)。

技能
majiayu000

Kotlin Mcp Server Generator

100

Generate a complete Kotlin MCP server project with proper structure, dependencies, and implementation using the official io.modelcontextprotocol:kotlin-sdk library.

技能
github