Systematic debugging
Consult before attempting a fix when root cause or behavior is unclear, failures repeat, components interact, or guessing would be unsafe.From its SKILL.md
npx -y skills add BB-84C/superpowers-lite --skill systematic-debuggingAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.9 KB, as published. Nobody here has run it
Systematic Debugging
Root cause, evidence, reproduction, and semantic verification prevent plausible patches from becoming recurring failures. Use this workflow when root cause or behavior is genuinely unclear, a failure repeats, components cross a boundary, or risk makes a guess unsafe. A known, determined small repair may proceed directly with suitable verification.
Investigation lenses
The following are composable investigation lenses, not mandatory consecutive phases: observe/reproduce the meaningful symptom; inspect recent state and relevant boundaries; compare a working analogue; trace input or control flow; form a single hypothesis; run the smallest discriminating experiment. Select the lenses that reduce the uncertainty at hand and preserve the evidence.
Use a failing automated test when there is a meaningful seam. Otherwise use the strongest suitable reproduction, counterfactual, or readback. After a root fix, run semantic verification of the affected behavior and relevant regressions.
Escalation and containment
When evidence shows shared state, hidden coupling, or each attempted fix moves the symptom, reassess the architecture before stacking more local changes. This is evidence-driven, not a fixed failure count.
A clearly labeled containment or mitigation may reduce harm while root cause is still under investigation. It must state its limitation and must not be presented as a root fix.
Supporting references
root-cause-tracing.mdfor tracing a deep symptom to its original trigger.defense-in-depth.mdfor safeguards across distinct trust boundaries and failure modes.condition-based-waiting.mdfor replacing arbitrary timing with observed conditions.