Skip to content
Back to skills

Simple Case Gate

ASecurity

Gate proposed research mechanisms, proof strategies, ansatzes, and universal identities on the simplest admissible nondegenerate case before testing harder cases; diagnose, repair, or stop when that case fails.

  • 50 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added September 3, 2026
ai-agentsgotesting

Security analysis

A100/100

Scanned September 3, 2026

npx -y skills add Mexregkan/claude-for-researchers --skill simple-case-gate --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Simple Case Gate?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Simple Case Gate
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/mexregkan-simple-case-gate/badge)](https://www.skillsdirectory.com/skills/mexregkan-simple-case-gate)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: simple-case-gate
description: Gate proposed research mechanisms, proof strategies, ansatzes, and universal identities on the simplest admissible nondegenerate case before testing harder cases; diagnose, repair, or stop when that case fails.
---

# Simple-case gate

Use this whenever proposing or evaluating a mathematical or physical mechanism,
proof architecture, ansatz, recursion, or claim intended to hold across a family.

## Non-negotiable rule

Identify the simplest admissible nondegenerate case and make the exact proposal work
there before increasing order, weight, depth, rank, spin, loop number, or the number of
free parameters. A genuine failure in that case forbids trying harder cases in the hope
that added complexity will repair a proposal that was claimed to work in every case.

## Gate

1. State the proposal and its domain precisely. If a simple case is excluded, derive the
   exclusion from the proposal rather than adding it after seeing a failure.
2. Choose the lowest-complexity case inside that domain. Verify that it is nondegenerate
   and that the test is sensitive to the proposed mechanism.
3. Test the complete proposal, with the same definitions, normalization, and conventions
   intended for the general case. Preserve the first exact residual or counterexample.
4. If it fails, stop escalation. Diagnose whether the cause is an implementation error,
   convention mismatch, degeneracy, false auxiliary assumption, or failure of the idea.
5. Either modify the proposal for a principled reason that explains the residual and
   retest the modified proposal on the same simple case, or abandon/narrow the claim.
6. Proceed to harder cases only after the simple gate passes. A modified proposal is a
   new proposal and must pass the gate again from the beginning.

## Forbidden responses to failure

- Moving directly to a higher order, weight, depth, rank, spin, or loop number.
- Adding parameters or correction terms available only in the harder case without first
  deriving why they must already resolve the simple residual.
- Calling the failed case “too simple” unless it is proved to lie outside the stated domain.
- Replacing the exact failed gate with a looser numerical, projected, or averaged test.

Report the gate case, why it is admissible and nondegenerate, the exact result, and what
the result licenses. A pass licenses the next case; it does not prove the general claim.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.