deep-research

Investigate undocumented behavior with source-referenced, confidence-graded findings.

3|1|Updated Dec 3, 2025
One-click install
npx skills add https://github.com/nsheaps/ai-mktpl --skill deep-research-nsheaps
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/nsheaps/ai-mktpl/tree/main/.claude/skills/deep-research
Command: npx skills add https://github.com/nsheaps/ai-mktpl --skill deep-research-nsheaps

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a repeatable, evidence-driven methodology for investigating undocumented behavior, internal mechanisms, and "how does X work?" questions when official documentation and community sources are incomplete or contradictory. It helps researchers move from uncertain observations to verifiable explanations with source references and confidence judgments.

Core Features & Use Cases

  • Multi-phase workflow: External research for broad coverage, source analysis for definitive answers, and synthesis with confidence levels and corrections.
  • Source tracing techniques: Guidance for locating entry points, tracing call chains, and working with decompiled or minified code to verify behavior.
  • Practical outputs: Produces research reports saved to disk, annotated findings with file:line references where available, and explicit remediation or usage guidance.
  • Use Case: Investigate why a CLI or UI emits unexpected titles or telemetry by combining GitHub issues, docs, and decompiled code to produce a final, source-backed report.

Quick Start

Run a deep research session on "terminal title behavior" and produce a source-backed report with confidence levels and file:line references.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I investigate undocumented behavior in decompiled binaries or community plugins?

Investigate undocumented behavior by applying a multi-phase research workflow that combines parallel external searches with optional source-level tracing and decompilation. This methodology produces clear, source-referenced explanations with file:line evidence.

What's the best way to trace an unexpected CLI telemetry issue back to its source code?

Trace unexpected CLI telemetry by combining GitHub issues, official documentation, and decompiled code to identify entry points and trace call chains. This produces a source-backed report with explicit confidence levels and file:line references.

Can I get confidence-graded findings when official documentation contradicts observed behavior?

Yes, you can obtain confidence-graded findings when documentation contradicts observed behavior by running a multi-phase research workflow. It synthesizes external searches and source analysis to deliver verifiable explanations with explicit confidence judgments.

How does source analysis work on minified or decompiled code to verify internal mechanisms?

Source analysis on minified or decompiled code works by locating entry points and tracing call chains to verify internal mechanisms. It provides definitive answers and annotates findings with file:line references where available.

When should I use reverse engineering and decompilation instead of relying on community documentation?

Use reverse engineering and decompilation when official documentation is insufficient or contradictory and community sources lack answers. It enables source-level tracing to move from uncertain observations to verifiable, evidence-driven explanations.