five-whys
Skill1 starsChanged 6 days agoApache-2.0View on GitHub
Run a Five Whys analysis (Toyota) with the person. It asks why, one step at a time, checks each answer, and walks each branch down to its root cause, then draws a clean chart with the measures. It works in the chat, and on the local whiteboard of this plugin (server board-control) when the person asks for it or agrees to it. Use when the user types /gr-whiteboard:five-whys, asks for a Five Whys (Nazenaze) analysis, or when consult-on-board or issue-driven picks it.
9 other projects use a version of this.
Adopted by
Most recently taken up by
- shawnclybor/clybor-claude-toolingtheir own version, first seen Nov 8, 2026
- RBraga01/a-teamtheir own version, first seen Nov 1, 2026
- kitloomlk-stack/testing-debugging-kittheir own version, first seen Oct 15, 2026
- ysskrishna/ai-agent-skillstheir own version, first seen Oct 13, 2026
- karozi/awesome-product-management-skillstheir own version, first seen Sep 5, 2026
- tupe12334/instincttheir own version, first seen Aug 17, 2026
Used with
Projects with it also use pre-mortem.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Ingramradical235Skillpre-mortem
Pre-mortem analysis, scenario planning, future state narratives, and contingency frameworks. Use before finalizing any Strategy Kernel, when planning major initiatives, or when preparing Plan B triggers.
Also in 3 projects that use this
Other versions9 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 9
- ysskrishna/ai-agent-skills★ 25same name
- RBraga01/a-team★ 24same name
- jackjin1997/ClawForge★ 12same name
- flpbalada/fb-skills★ 8same name
- ozten/skills★ 7same name
- shawnclybor/clybor-claude-tooling★ 6same name
- karozi/awesome-product-management-skills★ 2same name
- tupe12334/instinct★ 1same name
- kitloomlk-stack/testing-debugging-kit★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- consult-on-boardSkill
- debug-board-serverSkill
- four-formsSkill
- issue-drivenSkill
- talk-on-boardSkill
The file
We split this skill into the rules it gives an agent
9 rules. 0 are common, 8 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
8
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Ask why with the person, one step at a time, until each branch reaches its root cause.
- Very specific: line 14
After open_board, start the watch Monitor of talk-on-board (Open step 2, client.js watch) at once, or you miss every Send.
- Distinctive: line 21
Check each answer as you go.
- Distinctive: line 26
When two conditions must both hold, write them as two boxes under the same why, not as a chain.
- Distinctive: line 29
Do not skip a step.
- Distinctive: line 30
Do not stop at blaming a person.
- Distinctive: line 33
Do not say the same thing twice in other words.
- Distinctive: line 39
When all branches are done, ask "Shall I draw the clean chart?".
- Distinctive: line 40
Before drawing, read each branch both ways: from the problem down with "why?", and from the root cause up with "so".
“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
five-whys
Ask why with the person, one step at a time, until each branch reaches its root cause. Use the person's words, not your own guesses.
Talk in the chat first, one why at a time in short bullets. Open the board only when the person asks for it, or when a picture would make it clearer and you ask "Shall we sort this out on the whiteboard?" and they agree; then run it as the skill talk-on-board says.
Board
- After
open_board, start the watch Monitor oftalk-on-board(Open step 2,client.js watch) at once, or you miss every Send. - From the top: the title (the framework name and the subject), the root cause definition from
SKILL_DIR/data/labels.jsonin small gray text, then the chart (SKILL_DIR is the folder of this file). - Answer as in a phone chat:
write_answerputs your answer below and left of the person's last comment. When the person only wrote in the chart, put it under your previous answer instead (at). - Draw the draft in the shape of
SKILL_DIR/data/chart-ja.mmdfor a Japanese speaker, orSKILL_DIR/data/chart-en.mmd, with the person's words: the problem on the left, the whys to the right, occurrence on top and outflow below, each first branch marked at its head. When a branch grows, add its boxes to the right.
Checks
Check each answer as you go. When one fails, ask a short question; do not fix it yourself. When the subject is plain, add it yourself, and confirm such additions once, when a step settles.
- One box holds one sentence. Do not leave out the subject or the verb, nor the object of a transitive verb.
- One box holds one cause. When an answer has two causes, split the branch.
- The chain explains the specific part of the problem (where, when, which one), not only a general cause.
- When two conditions must both hold, write them as two boxes under the same why, not as a chain.
- The subject is a role (worker, designer), never a person's name.
- Each answer is a cause of the answer before it, not a measure.
- Do not skip a step.
- Do not stop at blaming a person. Ask what let the mistake happen.
- Occurrence (why it happened) and outflow (why it was not caught) are separate branches. Split other branches when needed.
- Five is not a rule. Stop at the root cause, sooner or later.
- Do not say the same thing twice in other words. Do not let the logic loop.
- A root cause is a cause that prevents the problem from coming back when it is removed, and that the team itself can act on with a concrete control or improvement.
Clean chart
- For each root cause, ask the person for a measure and settle it in one sentence. When the person has none, ask the people or files that know the facts, and offer candidates for an OK. When the person leaves it to you, fill it from those sources and mark the box as a proposal. Each measure removes its root cause and agrees with every why on its branch. Each measure comes from a root cause on the chart. For a measure with no root cause, add its branch first.
- When all branches are done, ask "Shall I draw the clean chart?". Draw only after a yes.
- Before drawing, read each branch both ways: from the problem down with "why?", and from the root cause up with "so". Each link holds both ways; where one does not, ask the person.
- Draw it right of the draft with
draw_mermaidin the shape of the template, one row per branch; keep the draft. Head each box with its step ("Why 1", "Why 2", and "Root cause" for the root cause), in the person's language. Write its own title and the gray definition above it. - Line up the root causes in one column and the measures in the next (
edit_items). - Make the root cause boxes thick (
edit_items,strokeWidth4).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

