debug-mode

Reproduce and diagnose software bugs with instrumented logging and committed repro scripts.

6|Updated Feb 23, 2026
One-click install
npx skills add https://github.com/franzenzenhofer/debug-mode-skill --skill debug-mode-franzenzenhofer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug-mode
Source: https://github.com/franzenzenhofer/debug-mode-skill/tree/main
Command: npx skills add https://github.com/franzenzenhofer/debug-mode-skill --skill debug-mode-franzenzenhofer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Debug Mode eliminates guesswork by turning ad hoc debugging into a disciplined, evidence-driven workflow that reproduces failures, captures targeted runtime data, and produces verifiable fixes and regression tests. It prevents wasted time on blind fixes, reduces fragile patches, and enforces repeatable reproduction before any code change.

Core Features & Use Cases

  • Hypothesis-driven triage: Generate 3–5 ranked theories with clear evidence requirements for each.
  • Commit-first reproduction: Create and commit a failing repro script or test so instrumentation can be added and safely removed.
  • Instrumented logging & analysis: Add hypothesis-tagged [DEBUG-MODE] logs, capture to a file, and grep-filter results to protect context and enable precise analysis.
  • Safe cleanup and verification: Use git restore to remove instrumentation, apply minimal root-cause fixes, and perform mandatory red-to-green verification with regression tests.

Quick Start

Ask the assistant to "Enter Debug Mode: create and commit a failing repro, instrument suspect paths with [DEBUG-MODE] logs, capture and analyze debug.log, reset instrumentation with git restore ., implement the fix, and verify red-to-green."

Frequently Asked Questions about debug-mode

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

FAQPage Schema
How do I systematically reproduce and fix flaky race conditions in my code?

To reproduce and fix flaky race conditions, create a committed failing reproduction script, instrument suspect paths with tagged logs, capture output to a file, and analyze the logs to isolate the root cause. Apply a minimal fix and verify with a regression test.

What is the best way to debug unit test failures without guessing the root cause?

The best way to debug unit test failures is using a hypothesis-driven triage approach. Generate ranked theories with evidence requirements, add targeted instrumentation logs, grep the captured debug file to confirm the hypothesis, and apply a minimal root-cause fix.

How can I safely add and remove debugging logs when diagnosing API errors?

To safely add and remove debugging logs for API errors, commit a failing repro script first, add hypothesis-tagged logs on an isolated git branch, then use git restore to cleanly remove all instrumentation after analyzing the captured logs.

Does this debugging workflow require an isolated git branch for capturing runtime data?

Yes, this debugging workflow requires an isolated git branch and stashes. It uses commit-first reproduction to safely add and remove instrumentation logs, ensuring runtime data capture remains isolated and code cleanup via git restore is verifiable.

How do I verify a bug fix works and prevent future regressions after debugging?

To verify a bug fix and prevent regressions, perform mandatory red-to-green verification. Run the committed failing reproduction script or test, implement the minimal fix, and confirm the test passes from a failing red state to a green passing state.