deep-dive
Investigates root causes and defines requirements through a structured 2-stage pipeline.
Install
mkdir -p .claude/skills/deep-dive && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/14940" && unzip -o skill.zip -d .claude/skills/deep-dive && rm skill.zipInstalls to .claude/skills/deep-dive
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
2-stage pipeline: trace (causal investigation) then deep-interview (requirements). Activate when user says: deep dive, deep-dive, investigate deeply, trace and interview.Key capabilities
- →Trace the causal factors of a problem
- →Conduct a deep interview to define requirements
- →Generate hypotheses for tracing investigations
- →Execute parallel tracer lanes to gather evidence
- →Inject trace findings into the interview process
- →Converge on findings through rebuttal rounds
How it works
The skill orchestrates a two-stage pipeline: first, it traces the cause of a problem by generating and confirming hypotheses across parallel lanes, then it conducts a deep interview, injecting the trace findings to define requirements.
Inputs & outputs
When to use deep-dive
- →Debug root causes
- →Investigate feature requirements
- →Explore system bugs
- →Trace causal factors
About this skill
Deep Dive
Orchestrates a 2-stage pipeline: first investigate WHY something happened (trace), then define WHAT to do about it (deep-interview). Trace findings feed into the interview via 3-point injection.
Pipeline
deep-dive → ralplan (consensus refinement) → autopilot (execution)
When to Use
- User has a problem but doesn't know the root cause
- Bug investigation: "Something broke and I need to figure out why"
- Feature exploration: "I want to improve X but first need to understand it"
When NOT to Use
- Already know the root cause → use
/deep-interview - Clear specific request → execute directly
- Investigation only, no requirements → use
/trace
Phases
Phase 1: Initialize
- Parse problem, detect brownfield/greenfield
- Generate 3 trace lane hypotheses (code-path, config/env, measurement/artifact)
Phase 2: Lane Confirmation
Present hypotheses to user for confirmation (1 round).
Phase 3: Trace Execution
Run 3 parallel tracer lanes using @tracer agents:
- Each lane: evidence for, evidence against, critical unknown, discriminating probe
- Rebuttal round between top hypotheses
- Convergence detection
- Save to
.omc/specs/deep-dive-trace-{slug}.md
Phase 4: Interview with Trace Injection
Follow deep-interview protocol with 3 overrides:
- initial_idea enrichment: Include trace's most likely explanation
- codebase_context replacement: Use trace synthesis (skip re-exploration)
- question queue injection: Per-lane critical unknowns become first questions
Low-confidence trace: don't inject uncertain conclusion, use ALL unknowns as questions.
Phase 5: Execution Bridge
Same options as deep-interview: ralplan → autopilot (recommended), autopilot, ralph, team, or refine further.
Output
Spec saved to .omc/specs/deep-dive-{slug}.md with additional "Trace Findings" section.
When not to use it
- →When the root cause is already known
- →When there is a clear and specific request that can be executed directly
- →When only an investigation is needed, without defining requirements
Limitations
- →Requires user confirmation for trace lane hypotheses
- →Low-confidence traces may lead to more questions in the interview
- →Relies on @tracer agents for trace execution
How it compares
This skill integrates causal investigation (tracing) with requirements gathering (deep interview) into a single pipeline, ensuring that solutions are grounded in a thorough understanding of the problem's root cause, which is more comprehens
Compared to similar skills
deep-dive side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| deep-dive (this skill) | 0 | 4mo | No flags | Advanced |
| deep-research | 35 | 9mo | Review | Advanced |
| recall-reasoning | 2 | 7mo | Review | Intermediate |
| math-progress-monitor | 1 | 7mo | No flags | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
deep-research
davidorex
Multi-agent parallel investigation for complex VCV Rack problems
recall-reasoning
parcadei
Search past reasoning for relevant decisions and approaches
math-progress-monitor
parcadei
Metacognitive check-ins during problem solving - detects when to pivot or persist
memory-grep
zxl85813-web
传统检索引擎集成：利用 SmartGrep (BM25 + Fuzzy) 召回向量匹配难以处理的精确事实。当用户询问特定错误 ID、唯一代号、或特定的代码 API 细节时触发。
analyse
eschaar
Cross-cutting technical analysis. Investigates impact, tradeoffs, root causes, or feasibility without implementing changes. Use when asked to "analyse this", "investigate the impact", "what are the tradeoffs", "root cause analysis", "is this feasible?", or "compare these approaches". Produces an ana
exploring-codebases
tmcfarlane
Semantic search for codebases. Locates matches with ripgrep and expands them into full AST nodes (functions/classes) using tree-sitter or pre-generated _MAP.md files. Returns complete, syntactically valid code blocks rather than fragmented lines. Use when looking for specific implementations, exampl