practice
Skill0 starsChanged 30 days agoMITView on GitHub
Practice product judgment on a decision case. Presents a case without the answer, takes the user's reasoning, Cynefin domain and call, grades it on the Gate, Reductionism, Systems thinking and Referee with specific feedback, then shows the worked tradeoff card. Use when the user wants to practice, drill or exercise product decisions, train product judgment, or test themselves on systems thinking, reductionism or Cynefin.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- tlifschitz/ieeextreme-preptheir own version, first seen Nov 1, 2026
- farazahmediu01/agents-sdk-fluencytheir own version, first seen Oct 15, 2026
- ccch1mneyyy/dsh-TUItheir own version, first seen Oct 2, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- ccch1mneyyy/dsh-TUI★ 7.8ksame name
- tlifschitz/ieeextreme-prep★ 2same name
- farazahmediu01/agents-sdk-fluency★ 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 repositoryOne other file it gives its agents.
- decideSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 29
If the user names a case, use it.
- Distinctive: line 30
If the user brings their own decision, use it.
- Distinctive: line 37
Ask for, in a few lines:
“Common” is measured: kept by at least 6 in 10 of the 3 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
Practice: train the two lenses and the referee
Reference files, relative to this SKILL.md (the same files are at ${CLAUDE_SKILL_DIR}/references/):
references/cases/README.md: the case index.references/cases/annual-discount.mdreferences/cases/onboarding-steps.mdreferences/cases/referral-reward.mdreferences/cases/support-chatbot.mdreferences/cases/daily-push.mdreferences/cases/take-rate.mdreferences/cases/cta-copy.mdreferences/cases/feature-parity.mdreferences/cases/partner-lead.mdreferences/rubric.md: how to grade.references/archetypes.md: the eight classic archetypes, for feedback on loops.
Each case file has "The situation", then a solution marker line, then the worked tradeoff card.
1. Pick the case
- If the user names a case, use it.
- If the user brings their own decision, use it. There is no stored solution for it.
- Otherwise, list the cases from the index, one line each with the decision only. Never show the domain or the leader. Let the user pick, or start with the first case not done in this conversation.
2. Present it without the solution
Show only "The situation" and the task line, nothing below the solution marker. Never reveal the domain, the leader, the call or any part of the card before the user answers, unless they ask for it.
Ask for, in a few lines:
- The need in the requester's words, and the input that decides the call.
- The direct effect, with a number or two and the arithmetic.
- At least one loop, with signs, and whether it is R or B.
- A Cynefin domain and who leads.
- One call: Go, Go with guardrails, Redesign, Don't, or Get X first.
3. Grade
Use references/rubric.md. Score the Gate from 0 to 3, and Reductionism, Systems thinking and Referee from 0 to 4 each. For a case that stops at the gate, score the Gate and the Referee's call line only. Quote the user's own line in each piece of feedback, say what is right or missing, and give the exact fix. Recompute the user's arithmetic and check their R or B against their signs. Do not take points off for a different number when the assumption is stated and the arithmetic checks.
4. Show the worked solution
Then show the worked card: everything below the solution marker. Point to the one or two places where the user's answer and the card differ most, and say which test made the referee decide as it did.
For the user's own decision, grade with the rubric, then build the worked card with the decide skill (/reductionism-vs-systems-thinking:decide) on the same decision.
5. Next
Offer the next case, or a harder variant: change one fact that would move the decision to another domain, and ask again.
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.

