skills.rest
INNERJOINT
INNERJOINTCOMMUNITY
Explore all skills by this author →
GitHub

deep-dive

Trace root causes and produce interview-driven specifications for downstream planning.

Updated Mar 26, 2026
Search Tags:#investigation#specification#requirements#orchestration#interview#tracing#root-cause
One-click install
npx skills add https://github.com/INNERJOINT/HarnessSkills --skill deep-dive-innerjoint
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-dive
Source: https://github.com/INNERJOINT/HarnessSkills/tree/main/skills/deep-dive
Command: npx skills add https://github.com/INNERJOINT/HarnessSkills --skill deep-dive-innerjoint

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep Dive investigates ambiguous, evidence-heavy problems by running a structured causal trace and a targeted deep-interview, then crystallizes the results into a reproducible, actionable specification for downstream planning and execution.

Core Features & Use Cases

  • Two-stage pipeline: a 3-lane trace followed by a deep-interview with a 3-point injection that seeds questions, context, and hypotheses.
  • Brownfield and greenfield handling: automatic exploration of existing codebases for brownfield cases and graceful degradation for low-confidence traces in exploratory work.
  • Artifact persistence and handoff: saves trace and spec artifacts to .omc/specs/ and provides explicit handoff to omc-plan, autopilot, ralph, or team for execution.

Quick Start

Run a deep dive by saying: /deep-dive "Describe your problem or exploration target".

Frequently Asked Questions about deep-dive

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

FAQPage Schema
How do I investigate root causes of ambiguous software problems?▼

Investigating root causes of ambiguous software problems requires running a structured causal trace and a targeted deep-interview to crystallize evidence-backed findings into a reproducible requirements specification for downstream execution.

What is the best way to turn code tracing findings into a requirements specification?▼

Transforming trace findings into a requirements specification involves a two-stage pipeline that applies a 3-point injection to seed questions and hypotheses, yielding a persisted spec artifact for downstream planning.

Can I use causal tracing for exploratory feature work in greenfield projects?▼

Causal tracing supports greenfield projects by applying graceful degradation for low-confidence traces during exploratory analysis, automatically adjusting its investigation depth to suit new codebase explorations.

How do I start a root-cause analysis on an existing codebase?▼

Starting a root-cause analysis on an existing codebase involves invoking the investigation command with a problem description, which automatically explores the brownfield codebase and initiates the causal trace pipeline.

Does the requirements specification integrate with downstream planning tools?▼

The requirements specification provides explicit handoff to omc-plan, autopilot, ralph, or team by persisting trace and spec artifacts to the .omc/specs/ directory, enabling seamless downstream planning and execution.

Related Skills

Explore More Software Engineering Skills→
inflexa-aiinflexa-ai
29

openspec-explore

Analyze codebase patterns and visualize system flows with OpenSpec CLI.

Official
Intermediate
Moliboy5000Moliboy5000

cancel

Cancel active OMC modes by detecting sessions and executing cleanup scripts.

Community
Advanced
officechbusinessservices-creatorofficechbusinessservices-creator

autopilot

Automate end-to-end software delivery from product idea to working codebase.

Community
Advanced
OliverOuyangOliverOuyang
1

deep-dive

Trace ambiguous system behavior and produce evidence-backed requirements specifications.

Community
Advanced