# Notagenius

> Turn a raw product or service idea into an evidence-backed, economically plausible service through JTBD discovery, competing bets, experiments, value-path design, outcome measurement, and falsification. Use when deciding whether an idea deserves investment, what evidence to gather next, or whether to continue, revise, return, or stop.

- Skill: `closedloop-technologies/notagenius` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds@latest add closedloop-technologies/notagenius`
- Raw SKILL.md: https://api.skillmd.com/api/skills/closedloop-technologies/notagenius/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: closedloop-technologies (https://skillmd.com/u/closedloop-technologies)
- Updated: 2026-09-10
- Page: https://skillmd.com/skills/closedloop-technologies/notagenius

---


# Not a Genius

Reduce consequential uncertainty before increasing implementation. Treat the
initial idea as one proposed solution, not as the object to prove. Stopping is
a valid outcome when evidence does not justify further investment.

## Choose the operating mode

- Read [PLAYBOOK.md](references/PLAYBOOK.md) when defining, reviewing, or
  applying the portable validation methodology. It is authoritative for what
  the artifacts, evidence standards, gates, and recursive transitions mean.
- Read [PLAN.md](references/PLAN.md) when executing the playbook with
  ClimbHill. It maps the methodology onto skills, Runs, Evidence Snapshots,
  recommendations, authority checks, and venture artifacts.

For an ordinary request, load only the document needed for the current mode.
When ClimbHill execution requires interpreting a playbook gate, read both.

## Core rules

- Judge progress by reduced uncertainty, not documents, cards, features, or
  code completed.
- Separate observations, inferences, assumptions, and contradictory evidence.
- Prefer observed behavior and direct measures. Match evidence strength to the
  consequence of the decision.
- Precommit experiment thresholds and outcome interpretations before exposure.
- Generate materially different bets before converging.
- Preserve stable lineage from claims through evidence and decisions to every
  feature, price, and implementation commitment.
- Design consequential external actions first. Obtain authorization before
  outreach, spending, production exposure, data access, or similar mutations.

## Skill composition

Use `$reasoning` as the evidence-oriented router. Its Discover, Bet, Smoke Test,
Value Path, and Falsify workflows are referenced guidance rather than separate
skills; its Experiment and Evidence documents are the core primitives. Use
`research`, `prototype`, `domain-modeling`, or `grilling` only when their
information source matches the active uncertainty.

Use `$business-soul` when service boundaries, accounts, customization, hosting,
pricing, trust, or go-to-market are in play. Keep the workflow spine open and
locally useful; charge for workload, workspace, warranty, or another concrete
cost or trust surface.

`wayfinder` and `ask-matt` are optional integrations. Use them only when they
are available: Wayfinder may map a decision space too large for one useful
session, and an engineering router may accept an evidence-justified MVP after
this workflow reaches its implementation handoff.

## Completion

End each uncertainty-reduction loop with the hypothesis evaluated, artifact and
Evidence IDs updated, the justified evidence status, an explicit venture
decision, and the single highest-value remaining uncertainty. Continue beyond
the next decision gate only when the user has authorized continuation.