systematic-debugging
Four-phase debugging framework that ensures root cause investigation before attempting fixes. Never jump to solutions. Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
74
76%
Does it follow best practices?
Impact
68%
0.85xAverage score across 2 eval scenarios
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./plugins/systematic-debugging/skills/systematic-debugging/SKILL.mdEvaluation results
63%
-17%Fixing a Silent Data-Loss Bug in a Utility Library
Test-first bug fix with single targeted change
Test file present
100%
100%
Bun test command
50%
40%
Test written first
0%
0%
Falsy-key test case
100%
100%
Root cause hypothesis
100%
100%
Single-line fix
100%
0%
Correct fix applied
100%
100%
No unrelated changes
100%
90%
74%
-7%Sales Pipeline Revenue Discrepancy
Systematic debugging of silent wrong output in a multi-stage Python pipeline
Investigation before fix
100%
100%
Diagnostic instrumentation added
0%
0%
Reader output boundary examined
62%
25%
Transformer boundary examined
87%
37%
Transformer identified as failing component
100%
100%
Specific hypothesis formed
100%
100%
Correct fix applied
100%
100%
Single file modified
100%
100%
Fix documented in log
100%
100%
Corrected report output
100%
100%
- Repository
- secondsky/claude-skills
- Commit
a0994f7
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.