investigator

Generate ranked hypotheses and track root cause analysis in a living investigation document.

Updated Jan 31, 2026
One-click install
npx skills add https://github.com/Kibbey/fyli --skill investigator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigator
Source: https://github.com/Kibbey/fyli/tree/main/.claude/skills/investigator
Command: npx skills add https://github.com/Kibbey/fyli --skill investigator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables teams to perform structured, hypothesis-driven root cause analysis by generating ranked hypotheses and maintaining a living investigation document that records hypotheses, tests, findings, and resolution status.

Core Features & Use Cases

  • Hypothesis generation and ranking: Produces a prioritized list of plausible causes with rationale and test plans.
  • Interactive test selection: Uses AskUserQuestion to let users choose which hypothesis to test first.
  • Living investigation doc: Automatically documents problem statements, tests, findings, and progress in docs/investigations/YYYY-MM-DD-<slug>.md.
  • Phase-guided workflow: Guides the user through problem definition, hypothesis testing, and resolution, with status tracking.

Quick Start

Create an investigation by defining the problem and creating an investigation document, then let the skill propose hypotheses and prompt you to pick one to test.

Frequently Asked Questions about investigator

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

FAQPage Schema
How do I perform structured root-cause analysis for complex debugging?

Structured root-cause analysis automates hypothesis-driven investigation by generating ranked causes with test plans. It maintains a living investigation document to track tests, findings, and resolution status across the debugging lifecycle.

What is hypothesis-driven investigation and how does it work?

Hypothesis-driven investigation generates prioritized plausible causes with rationale and test plans. You interactively select which hypothesis to test first, and results are documented automatically in a markdown file under docs/investigations.

How do I start an investigation and track my findings?

To start an investigation, define the problem statement to create an investigation document. The workflow then proposes hypotheses and prompts you to select one to test, tracking progress through problem definition, testing, and resolution phases.

Can I use this for incident reviews and complex problem-solving tasks?

Yes, hypothesis-driven root cause analysis applies to incident reviews and complex problem-solving tasks. It provides structured hypothesis testing and traceable documentation to guide you from problem definition through to resolution.

Where are investigation documents stored and what format are they?

Investigation documents are stored as markdown files at docs/investigations/YYYY-MM-DD-<slug>.md. This living document automatically records problem statements, tests, findings, and progress throughout the investigation lifecycle.

Do I need any dependencies or components to run an investigation?

No dependencies or components are required to run an investigation. The workflow uses the AskUserQuestion tool to interactively select hypotheses and automatically stores progress in markdown format.