scope-challenge
Skill4 starsChanged 3 months agoMITView on GitHub
Capture a feature request as a very short PRD, then challenge it for the smallest valuable scope before implementing. Use on every functional or feature request — "add X", "we need Y", "build Z" — not only when the user asks to simplify, cut scope, or reduce the MVP.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- matthewdigiuseppe/MStacktheir own version, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- matthewdigiuseppe/MStack★ 17same 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 repository19 other files it gives its agents.
- AGENTS.mdAGENTS.md
- adapter-contract-testingSkill
- azure-devops-prSkill
- bugszero-root-causeSkill
- legacy-3pSkill
- port-refactoringSkill
- quality-gateSkill
- review-ai-writingSkill
- review-allSkill
- review-architectureSkill
- review-complexitySkill
- review-driftSkill
- review-impactSkill
- review-prSkill
- review-securitySkill
- review-slopSkill
- review-testingSkill
- teams-local-messagesSkill
- write-testsSkill
The file
Files
Scope Challenge
A small codebase is always easier to change than a large one. Every functional request is therefore challenged on its simplicity before implementation, not after.
This runs at the front of docs/feature-workflow.md: it
questions what is being built, then hands the agreed scope to the workflow, whose step 2
slices how it gets delivered. Challenge the scope first — slicing a request that should
have been half its size just produces more slices.
Steps
- Capture a very short PRD as an artefact of the session: the need, in a handful of lines, in domain terms. No implementation detail. If it does not fit on a screen, it is describing a solution rather than a need.
Look for a higher-value subset, along one or more of these axes:
- Shrink the problem. Is a subset of the functional cases enough for the next step?
- Shrink the population. Is a subset of the users enough — one team, one tenant, one region — with the rest following once it has proved out?
- Hand the edge case back to a human. Can a rare case be left to fail explicitly, with a clear exception, instead of building the logic that handles it? An edge case that occurs twice a year rarely repays the branch that handles it, and the branch is never exercised.
- Offer the user several scope reductions, each with its trade-off, and let them choose. Do not pick for them, and do not proceed to implementation on your own reading — the point of the exercise is the decision, not the recommendation.
- Update the PRD to the agreed scope. It stays very short.
When to Stop
Say so when the request is already minimal. A request that cannot usefully be cut is the normal case for a bug fix, a one-rule change, or a slice the user has already narrowed — report that in a line and move on. Manufacturing three options for a two-hour change wastes more than it saves.
Why
- It makes the next step to implement unambiguous, which is most of what a small scope buys.
- The short, current PRD makes any gap between the need and the implementation visible —
which is what a later
review-complexitypass needs in order to call code accidental rather than merely unfamiliar.
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.

