At a glance
- Best for
- Investigate root causes in stages
- Works with
- Errors, reproduction steps, and relevant code · Recent changes and existing diagnostic data
- Outcome
- Root causes and supporting evidence · A repair direction or implementation
- Source
- obra/superpowers
01 / Understand the skill
About systematic-debugging
Find root causes through reproduction, evidence gathering, pattern comparison, and single-hypothesis experiments, then verify fixes with regression tests. Suitable for failing tests, unexpected behavior, and repeated unsuccessful fixes. Takes errors, recent changes, and reproduction steps and returns root-cause evidence, a repair direction, and regression results, avoiding unverified patch stacking.
Requirements
- An agent that can read errors, code, and test results
- The companion test-driven-development and verification-before-completion skills
- The project's own environment and data permissions for running tests
Inputs
- Errors, reproduction steps, and relevant code
- Recent changes and existing diagnostic data
Outputs
- Root causes and supporting evidence
- A repair direction or implementation
- Regression-test and verification results
02 / Confirm the fit
Use systematic-debugging when you need to
- Investigate root causes in stages
- Minimal hypothesis experiments and an evidence trail
- Regression testing and fix verification
03 / See the workflow
How systematic-debugging works
- 01Reproduce the failure and gather evidence
- 02Compare working paths and test a root-cause hypothesis
- 03Run regression tests and relevant checks after the fix
Check before installing
Compatible platforms
Cursor · GitHub Copilot · OpenAI Codex · Claude Code
Best fit
Investigate root causes in stages
Review the source instructions if the required output differs from the formats described above.
Install and use
Compatible with Cursor and GitHub Copilot and OpenAI Codex and Claude Code · Published by obra
Ready to add systematic-debugging?
Use the published instructions below, then provide the required input and describe the result you need.
General instructions · Cursor · GitHub Copilot · OpenAI Codex · Claude Code
/plugin install superpowers@claude-plugins-official Install the complete Claude Code plugin as described in the upstream README. For other hosts, follow their dedicated installation sections in the same README.
Review the source repository and license before installing third-party skills.
Continue exploring
Similar skills
Authoritative reference for the Claude API and Anthropic SDKs: model IDs, pricing, thinking/effort params, streaming, tool use, caching, managed agents, and model migration.
Read, edit, create, and fix Excel/CSV files — formulas, formatting, charts, and clean financial models with zero formula errors.
Generate distinctive, production-grade frontend UI that avoids generic AI aesthetics — pages, components, dashboards, and landing pages.
Build high-quality MCP (Model Context Protocol) servers in Python (FastMCP) or Node/TypeScript that let LLMs use external services through well-designed tools.