One-click install
npx skills add https://github.com/nichobbs/lyric-lang --skill deep-dive-nichobbs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-dive
Source: https://github.com/nichobbs/lyric-lang/tree/main/.claude/skills/deep-dive
Command: npx skills add https://github.com/nichobbs/lyric-lang --skill deep-dive-nichobbs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep-dive turns ambiguous issues into evidence-based requirements by first investigating “why” and then guiding an interview to define “what to do” with the trace findings carried forward.

Core Features & Use Cases

  • 2-stage investigation to spec: Runs a causal trace (3 parallel lanes) and then performs a targeted requirements crystallization interview.
  • 3-point trace injection: Enriches the interview’s starting problem, reuses trace-derived context, and seeds the first interview questions from per-lane critical unknowns.
  • Resume-ready artifacts: Persists both trace and final spec to .omc/specs/ with state fields designed for interruption/resume robustness.

Quick Start

Run deep-dive on your problem description to get a requirements spec backed by a causal trace, then hand it off to planning and execution when ready.

Frequently Asked Questions about deep-dive

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn an ambiguous bug investigation into actionable requirements?

Root cause analysis transforms ambiguous bug investigation into actionable requirements by running a 2-stage trace-and-interview pipeline. It executes a causal trace with 3 parallel hypotheses to identify why an issue occurs before guiding an interactive interview to define what to do.

What is the best way to generate specs from evidence-based system investigation?

Spec generation from evidence-based system investigation uses ambiguity-threshold-gated logic to crystallize requirements. It injects safe trace-context from 3-point parallel evidence gathering into an interactive interview, ensuring the final spec is backed by confirmed causal hypotheses.

How does interview orchestration work for feature exploration requirements?

Interview orchestration for feature exploration seeds the first interview questions from per-lane critical unknowns. It reuses trace-derived context and enriches the starting problem description, targeting ambiguity resolution before generating any planning handoff specs.

Do I need to provide a complete problem description to start root cause analysis?

Root cause analysis does not require a complete problem description. You can run it on an ambiguous issue, and the pipeline will orchestrate parallel evidence gathering and an interactive interview to resolve root-cause uncertainty before generating any specs.

How do I hand off requirements specs to downstream planning and execution?

Planning handoff for downstream execution is handled via an explicit spec_path. The skill persists both the causal trace artifacts and the final requirements spec to the .omc/specs/ directory, using state fields designed for interruption and resume robustness.