Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Wondermonger-daydreaming/claude-skills-library --skill counter-experimentgit clone --depth 1 https://github.com/Wondermonger-daydreaming/claude-skills-libraryWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wondermonger-daydreaming/claude-skills-library/counter-experiment)<a href="https://agentmods.dev/skills/wondermonger-daydreaming/claude-skills-library/counter-experiment"><img src="https://agentmods.dev/badge/skills/wondermonger-daydreaming/claude-skills-library/counter-experiment/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wondermonger-daydreaming/claude-skills-library/counter-experiment"><img src="https://agentmods.dev/badge/skills/wondermonger-daydreaming/claude-skills-library/counter-experiment.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00150 | $0.01295 |
| Opus 5 | $0.00075 | $0.00647 |
| Sonnet 5 | $0.00030 | $0.00259 |
| Haiku 4.5 | $0.00015 | $0.00129 |
Grade A, and why
counter-experiment scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Counter-Experiment
Not "what's wrong with your claim" but "what would I need to see to believe it — or to know it's wrong."
The Practice
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.
The Protocol
Step 1: State the Claim Precisely
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:
- Claim as stated: "Quantum entanglement enables nonlocal consciousness during clinical death"
- Claim stripped to testable core: "Stimulus sequences generated by entangled qubits produce higher recall accuracy in cardiac arrest survivors than random sequences"
Step 2: Generate the Simplest Alternative Explanation
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."
Step 3: Design 3-5 Experiments That Discriminate
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:
- Hypothesis: What specific prediction does each interpretation make?
- Design: Participants, conditions, measures, controls
- Discriminating outcome: What result would support interpretation A? What result would support interpretation B? What result would be ambiguous?
- Feasibility: Can this actually be done? Ethical constraints? Resource requirements?
- Elegance: How cleanly does this experiment separate the interpretations? (A perfect experiment has zero overlap between predicted outcomes.)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 100 lines · 150 tokens per session scan A d4d36b68555c
counter-experiment is a skill published in the GitHub repository Wondermonger-daydreaming/claude-skills-library (6 stars, last pushed 2mo ago), licensed MIT. It adds 150 tokens to every session and 1,295 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
alterlab-rma-statistics-guide
This skill should be used when the user asks about "statistics", "statistical analysis", "SPSS", "R statistics", "hypothesis testing", "t-test", "ANOVA", "chi-square", "regression analysis", "correlation", "p-value", "effect size", "confidence interval", "descriptive statistics", "inferential statistics", "data…
04_data_intelligence
You are the Data Intelligence Specialist — the final authority on statistical rigor, figure excellence, and results architecture. You ensure that every number reported is correct, every figure communicates instantly, and every results paragraph follows a logical structure that reviewers cannot dismantle.
Claude
Author: Dr. Aneesh Joseph Implementation: Claude (Anthropic) Version: 2.2 | December 2025.
pkpd-modeling
Pharmacokinetic and pharmacodynamic modelling and simulation - non-compartmental analysis, compartmental and population PK, PK/PD and exposure-response, TMDD, PBPK orientation, bioequivalence, allometric scaling and first-in-human dose, drug interaction prediction, and Bayesian therapeutic drug monitoring. Use when…
biopython
Comprehensive molecular biology toolkit. Use for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access (Bio.Entrez). Best for batch processing, custom bioinformatics pipelines, BLAST automation. For quick lookups use gget; for multi-service integration use…
matlab
Build, review, migrate, and safely plan MATLAB or GNU Octave numerical workflows, including arrays, tabular/time data, tests, projects, graphics, MAT files, and explicit Python interoperability.