Skip to content

assess

skillnew

should this be built at all?

Evidence before investment. An idea ends in go, clarify or stop, with the reasons written down.

how it fires

You never type a command. Describe your problem in your own words and MasterMind recognizes it and applies this skill. Power users can invoke it directly with /assess.

the rule it routes on
Use when someone wants to know whether an idea deserves building before anyone builds it: "is this worth doing?", "should we build X?", "evaluate this idea", "would users actually want this?", "help me decide whether to invest in this", a product or business idea with no code yet, or a feature request nobody has said yes to. Not for scoping something already decided (that's `interview` or `specify`), not for adopt-vs-build on a library (that's `tech-scout`).

🧠 MasterMind ▸ assess

what this skill does

This is the real text MasterMind follows: generated from the source, not a summary of it.

The problem this solves

An AI will build anything you describe, which makes the most expensive question in product work disappear: should this exist at all? Ideas go straight from a sentence to a codebase, and the cost of a bad one is only discovered after it ships and nobody uses it.

How it actually works

MasterMind takes the idea through five stages, refining a single file, specs/ideas/<name>.md:

  1. Intake: the idea in one sentence, who asked, who it serves, and the problem it removes.
  2. Research: what people do today instead, competitors, and any earlier attempt. Every finding is cited and labelled observed, reported or assumed, so a guess never passes as a fact.
  3. Define: what would show it worked, as one to three measurable signals with targets and dates, plus the constraints it must live within.
  4. Shape: the smallest version that would test the outcome, and the two or three assumptions it rests on, each with the cheapest test that could prove it wrong.
  5. Decide: go, clarify or stop, with reasons and the evidence that would change the verdict.

Before deciding, the shaped idea goes to a reviewer in a fresh context that has not seen MasterMind’s reasoning, with instructions to attack the assumptions. The recommendation is MasterMind’s; the decision is yours, and the file records who made it.

It needs no code, so it works for a business idea as well as a feature request. A go hands the shaped version straight to specify.

When it fires

“Is this worth building?” “Should we add a marketplace to the app?” “Help me decide whether to invest in offline mode.”

🧠 MasterMind ▸ checking whether this idea earns a build
   └ assess · intake → research → define → shape → decide

When it does not fire

  • Something already decided. Pinning down what to build is interview; writing it up is specify.
  • Choosing a library or tool. Adopt-versus-build on a dependency is the tech-scout agent’s job.

What you get

One file with a clear verdict, the evidence behind it, the riskiest assumption and how to test it cheaply. A well-reasoned stop is a result too: it is the cheapest outcome an idea can have.