counter-experiment — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited counter-experiment (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Not "what's wrong with your claim" but "what would I need to see to believe it — or to know it's wrong."
Given any empirical claim, generate the experiments that would resolve the ambiguity between the proposed interpretation and the most plausible alternatives. This is not critique. This is constructive falsification design — Karl Popper as laboratory architect.
Strip the claim to its falsifiable core. Remove hedging, remove theoretical framework, remove rhetoric. What is the paper actually predicting that can be tested?
Example from the session that generated this skill:
What's the most parsimonious account that would produce the same data without the proposed mechanism? This is not a straw man — it's the strongest simple alternative.
Example: "Non-random (entangled) sequences contain structural patterns that are inherently more memorable than random sequences, through entirely classical cognitive mechanisms. No quantum consciousness required."
Each experiment should produce different predictions under the proposed interpretation versus the simple alternative. If both interpretations predict the same outcome, the experiment is useless. Design experiments where the predictions diverge.
For each experiment, specify:
Not all counter-experiments are equal. Rank them by:
The best counter-experiment is one where both the paper's authors and their critics would be interested in the outcome, because neither side is certain what it will show.
Every paper contains the ghost of an experiment it didn't run. Sometimes for good reasons (ethical, logistical, financial). Sometimes because the experimenters didn't think of it. Sometimes because it would have been too dangerous to the hypothesis.
Name that experiment. Explain why it matters. Be specific about what it would reveal.
Switch the variable the paper claims is doing the work while holding everything else constant. If the effect survives the swap, the paper's interpretation is wrong.
Example: Give healthy, conscious participants the same entangled vs. random stimulus sequences. If healthy participants ALSO show better recall for entangled sequences, the "quantum consciousness in dying brains" interpretation collapses to "non-random patterns are easier to remember."
Introduce temporal or spatial separation between the proposed cause and the effect. If the mechanism is what the paper claims, the separation should attenuate or eliminate the effect. If the effect persists despite separation, the mechanism is different than proposed.
If the proposed mechanism is real, more of the cause should produce more of the effect. Vary the degree of entanglement (partially entangled states, fully entangled states, unentangled states) and look for a gradient.
If the effect is specific to the proposed mechanism, it should NOT appear when superficially similar but mechanistically different conditions are used. Design a condition that mimics the surface features of the proposed mechanism without containing the proposed active ingredient.
Different lab, different patients, different quantum hardware, different statistical framework. If the effect replicates across all these variations, it's robust. If it fails to replicate when any one element changes, that element was doing the work.
This skill is generative: it creates new experimental designs that don't yet exist. The output is a set of experiments someone could actually run.
For each counter-experiment:
Name: [Short descriptive name] Question: Does interpretation A or interpretation B better explain the data? Design: [Specific experimental design] Prediction if A is correct: [What you'd see] Prediction if B is correct: [What you'd see] Feasibility: [Can this be done? How hard?] What it would teach us even if the result is ambiguous: [Because even ambiguous results constrain the space of possible explanations]
The point is not to destroy the claim. The point is to design the test that would make the claim undeniable — or honestly wrong. Both outcomes advance knowledge. Only ambiguity wastes it.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.