dig

Coordinate seeds, verification agents, and evidence synthesis for structured investigations.

5|Updated Mar 10, 2026

Search Tags:#investigation #codebase #verification #agents #dependency-mapping #root-cause-analysis #incident-response

Ready to run

npx skills add https://github.com/humblemuzzu/ghosttyyy --skill dig-humblemuzzu

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dig
Source: https://github.com/humblemuzzu/ghosttyyy/tree/main/pi-setup/config-skills/dig
Command: npx skills add https://github.com/humblemuzzu/ghosttyyy --skill dig-humblemuzzu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Structured investigations require coordinated exploration, evidence gathering, and verification to produce trusted conclusions.

Core Features & Use Cases

Epistemic standards and traceability: verify claims, label confidence, and falsify instead of confirming.
Phase-driven analysis: map seeds, explore dependencies forward and backward, and synthesize a complete picture.
Verification workflow: spawn verification agents for major claims and record VERIFIED, REFUTED, or INCONCLUSIVE results.
Synthesis and appendices: deliver structured findings with tables, trees, and appendices for auditability.
Autonomous discovery: discovery guided by seeds without hardcoded component lists, enabling bidirectional graph traversal.

Quick Start

Seed an inquiry, spawn verification-enabled agents, collect claims with file paths, and synthesize a final report.

Frequently Asked Questions about dig

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

FAQPage Schema

How do I perform root-cause analysis on a codebase with traceable evidence?▼

Root-cause analysis is performed by seeding an inquiry, spawning verification agents to explore dependencies bidirectionally, and synthesizing file-path evidence into a reproducible final report with appendices.

What is the best way to coordinate incident response investigations across complex dependencies?▼

Incident response investigations are coordinated by mapping seeds, traversing dependency graphs forward and backward, and verifying major claims to deliver structured findings with VERIFIED or REFUTED confidence labels.

Can I automate codebase archaeology without providing hardcoded component lists?▼

Codebase archaeology is automated through autonomous discovery guided by inquiry seeds, enabling bidirectional graph traversal and evidence synthesis without relying on hardcoded component lists.

How do verification agents label confidence during a structured investigation?▼

Verification agents label confidence by evaluating major claims against file-path evidence, assigning each claim a status of VERIFIED, REFUTED, or INCONCLUSIVE to ensure falsifiable conclusions.

Does dependency mapping support bidirectional traversal for incident response?▼

Dependency mapping supports bidirectional graph traversal, allowing forward and backward exploration from initial seeds to collect claims and synthesize a complete picture for incident response.

When do I need structured investigation with verification agents instead of manual codebase exploration?▼

Structured investigation with verification agents is needed when traceable, reproducible conclusions are required for incident response or root-cause analysis, replacing manual exploration with evidence-backed claim verification.