five-whys_skill
136
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill flpbalada/my-opencode-config --skill five-whys- SKILL.md7.8 KB
Overview
This skill conducts root cause analysis using the Five Whys technique to reveal underlying systemic issues. It guides teams through iterative questioning, documents evidence, and translates findings into prioritized, testable fixes. Use it to move from symptoms to durable solutions rather than temporary band-aids.
How this skill works
Start with a clear, measurable problem statement and ask "Why?" repeatedly, recording an evidence-backed answer at each step. Continue until the chain reveals a systemic root cause; extend beyond five whys if necessary. Capture participants, timestamps, and supporting data, then convert the root cause into prioritized solutions and success metrics.
When to use it
- Investigating recurring incidents or defects
- Debugging production failures or software bugs
- Analyzing customer churn or retention drops
- Conducting post-mortems or incident reviews
- Evaluating process delays, budget overruns, or quality issues
Best practices
- Define the problem precisely and with measurable terms
- Use data and evidence for each answer; avoid assumptions
- Involve diverse stakeholders to reduce blind spots
- Focus on process and system fixes, not individual blame
- Document the full why chain and validate the root cause with tests
Example use cases
- Production API returning 500 errors — find missing migration steps and add checks
- Customer churn spike — trace onboarding gaps and add automation
- Manufacturing stoppage — discover preventive maintenance lapses and implement schedules
- Project schedule slip — uncover dependencies and improve handoff processes
- Post-mortem for a major outage — convert root causes into prioritized remediation plan
FAQ
Five is a guideline; continue until you reach a verifiable systemic cause. Stop when the answer points to a correctable process or policy, not a person.
What if multiple root causes appear?
Document parallel chains and prioritize fixes by impact and effort. Use complementary methods like fishbone or Pareto analysis to surface and rank multiple causes.