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 hdu-ailab/EasyResearch --skill statistical-powergit clone --depth 1 https://github.com/hdu-ailab/EasyResearchWrote 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/hdu-ailab/easyresearch/statistical-power)<a href="https://agentmods.dev/skills/hdu-ailab/easyresearch/statistical-power"><img src="https://agentmods.dev/badge/skills/hdu-ailab/easyresearch/statistical-power.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 45 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00035 | $0.01251 |
| Opus 5 | $0.00017 | $0.00626 |
| Sonnet 5 | $0.00007 | $0.00250 |
| Haiku 4.5 | $0.00003 | $0.00125 |
Grade A, and why
statistical-power 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 7d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Statistical Power
Scope
Power is a design input, not a retrospective quality score. Use this Skill in the dispatch-selected experiment root before formal runs when sample size, minimum detectable effect (MDE), precision, clustering, attrition, multiplicity, or a nonstandard analysis makes evidence adequacy consequential.
Formal ML seed policy and power answer different questions. Five seeds estimate run-to-run variation; they do not establish dataset/sample size or detectable effect.
Record assumptions and accepted results in
<experiment-root>/formal-experiment-plan.md. Raw calculations, curves, and
sensitivity tables go under <experiment-root>/outputs/power/; accepted formal
copies may be promoted to <experiment-root>/results/power/. Power curves are
diagnostic evidence, not final manuscript figures.
Required Inputs
Derive or return blocked through the caller:
- primary outcome/metric and planned statistical test/model;
- unit of analysis and independent sample/cluster definition;
- effect-size or precision rationale with provenance;
- variance, event/base rate, ICC, attrition, allocation, and correlation assumptions when applicable;
- alpha, target power, sidedness, multiplicity, and stopping policy;
- feasible sample/compute range and sensitivity scenarios.
Never ask the user directly. Never choose an optimistic effect size only to make the feasible sample appear adequate. Report a range when assumptions are weak.
Procedure
- Match the power model to the exact planned analysis and experimental unit.
- Prefer pilot data, prior comparable studies, domain-relevant effect sizes, or an explicitly decision-relevant MDE. Record uncertainty and transport limits.
- Calculate required sample size and achieved power/MDE at feasible sizes.
- Inflate for clustering/design effect, repeated measures, attrition, unusable samples, imbalance, and multiplicity when the design requires it.
- Run sensitivity analysis across plausible effect/variance/rate assumptions.
- Use Monte Carlo simulation for mixed models, interactions, nonstandard estimators, adaptive rules, or analyses without a defensible closed form.
- Simulate the exact data-generating process and exact planned analysis. Use a fixed seed and report Monte Carlo uncertainty.
- Bound simulation work before launch and never tune assumptions after seeing target results without labeling the analysis exploratory.
- Record formulas/packages/versions, assumptions, output paths, and the chosen design consequence in the formal plan and experiment record.
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 7d ago First seen · 135 lines · 35 tokens per session scan A 2af6df42986f
statistical-power is a skill published in the GitHub repository hdu-ailab/EasyResearch (11 stars, last pushed 3d ago), licensed MIT. It adds 35 tokens to every session and 1,251 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
ai-research-reproduction
Rigor Reproduce compatible skill slug for README-first deep learning repository reproduction. Use when the user wants an end-to-end, minimal-trustworthy flow that reads the repository first, selects the smallest documented inference or evaluation target, coordinates intake, setup, trusted execution, optional trusted…
ai-research-explore
Rigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of currentresearch with auditable repo…
explore-run
Rigor Improve / Rigor Explore run leaf skill for bounded exploratory evidence in deep learning research repositories. Use when the researcher explicitly authorizes exploratory runs such as small-subset validation, short-cycle guess-and-check, batch sweeps, idle-GPU search, or quick transfer-learning trials, with…
env-and-assets-bootstrap
Rigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full…
analyze-project
Rigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrypoints, review configs and insertion points, or flag suspicious implementation patterns without modifying code or…
safe-debug
Rigor Debug / Rigor Audit skill for deep learning research work. Use when the user pastes a traceback, terminal error, CUDA OOM, checkpoint load failure, shape mismatch, NaN loss symptom, or training failure and wants conservative diagnosis before any patching, with debug fixes clearly separated from research…