investigate

Trace code paths and produce structured reports with risks.

19|2|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/bjornjee/agent-dashboard --skill investigate-bjornjee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/bjornjee/agent-dashboard/tree/main/adapters/claude-code/skills/investigate
Command: npx skills add https://github.com/bjornjee/agent-dashboard --skill investigate-bjornjee

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you understand a confusing behavior, failure, or architectural concern by tracing how the system works, without introducing changes that could worsen the issue.

Core Features & Use Cases

  • Read-only investigation: Produces conclusions without editing, writing, or creating files.
  • Structured codebase research: Traces code paths, reviews tests, and inspects configuration and dependencies.
  • Actionable reporting: Delivers findings, risks, and recommended next steps with concrete follow-up suggestions.

Quick Start

Ask the investigate skill to investigate a specific issue by stating your question or concern.

Frequently Asked Questions about investigate

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

FAQPage Schema
How do I investigate a codebase failure without making any code changes?

Codebase analysis for architectural concerns involves tracing call paths, reviewing tests, and inspecting configuration without modifying files. It enforces a read-only workflow to deliver line-referenced findings, identified risks, and recommended next steps.

Can I trace code paths and dependencies to understand a confusing system behavior?

Yes, tracing code paths and dependencies clarifies confusing behavior by mapping runtime assumptions across source and tests. This read-only investigation delivers line-referenced findings, identified risks, and recommended next steps without modifying files.

What is the best way to analyze git history and configuration for a risk assessment?

The best way to perform a risk assessment is a structured codebase research process that inspects git history and configuration. It traces behavior across source and tests to identify risks and recommend next steps without altering the codebase.

Does a read-only debugging investigation work for deep-dive performance and dependency tracing?

Read-only debugging works for deep-dive performance and dependency tracing by switching to main and pulling latest before analyzing source and tests. It enforces zero code changes while delivering structured reports with concrete follow-up suggestions.

What are the limitations of using a read-only workflow for codebase investigation?

The main limitation of a read-only codebase investigation is that it cannot directly resolve the issue; it only produces a structured report with findings and recommended next steps. All identified risks must be addressed in a separate workflow.