investigate

Debug Tumai platform failures through evidence-based root-cause analysis.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/mariasebarespersona/Tum-AI --skill investigate-mariasebarespersona
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/mariasebarespersona/Tum-AI/tree/main/.claude/skills/investigate
Command: npx skills add https://github.com/mariasebarespersona/Tum-AI --skill investigate-mariasebarespersona

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Systematic debugging with root-cause investigation. Four phases — investigate, analyze, hypothesize, fix. Iron Law — no fixes without root cause. Use when asked to "debug this", "fix this bug", "why is this broken", "investigate this error", or "root cause analysis". Proactively invoke this skill (do NOT debug directly) when the user reports errors, stack traces, unexpected behavior, "it was working yesterday", or is troubleshooting why something stopped working. This includes build failures, runtime errors, flow execution failures, and UI bugs.

Core Features & Use Cases

  • Establish a repeatable debugging framework that forces root-cause identification before any fix.
  • Collect symptoms, read the code path, and reproduce deterministically to verify hypotheses.
  • Apply an evidence-based workflow to reduce whack-a-mole debugging and improve long-term stability.
  • Use in scenarios like build failures, runtime errors, UI bugs, and behavioral regressions.

Quick Start

Describe the issue with symptoms and logs, and let the skill perform a root-cause investigation.

Frequently Asked Questions about investigate

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

FAQPage Schema
How do I find the root cause of a runtime error or build failure?

Root-cause debugging identifies the true cause of runtime errors and build failures by enforcing an evidence-based workflow. It collects symptoms, analyzes code paths, and verifies hypotheses through deterministic reproduction steps before applying any fixes.

What is the best way to debug recurring bugs that keep coming back?

Systematic root-cause investigation stops recurring bugs by enforcing an evidence-based workflow with no fixes until the true cause is found. It replaces whack-a-mole debugging with deterministic reproduction and guided hypothesis testing to improve long-term stability.

How do I troubleshoot unexpected behavior when something was working yesterday?

Root-cause debugging handles "it was working yesterday" cases by collecting evidence and analyzing code paths to identify the true cause of the unexpected behavior. It enforces deterministic reproduction to verify hypotheses before any fix is applied.

How do I perform root-cause analysis from a stack trace?

Performing root-cause analysis from a stack trace requires an evidence-based debugging workflow. You collect the stack trace as evidence, analyze the relevant code paths, and test guided hypotheses to deterministically reproduce the failure before fixing it.

Can I fix a UI bug without first reproducing the error deterministically?

No, you should not fix a UI bug without deterministic reproduction. Root-cause debugging enforces an Iron Law prohibiting fixes without identifying the true cause, requiring evidence collection and deterministic reproduction steps to verify hypotheses for UI bugs and behavioral regressions.

Does evidence-based debugging work for both UI bugs and build failures?

Yes, evidence-based debugging works for UI bugs, build failures, runtime errors, and behavioral regressions. It applies a four-phase framework—investigate, analyze, hypothesize, fix—to systematically identify the root cause across different types of unexpected behavior in the platform.