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 LeoLin990405/r-analytics-skill --skill h2ogit clone --depth 1 https://github.com/LeoLin990405/r-analytics-skillWrote 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/leolin990405/r-analytics-skill/h2o)<a href="https://agentmods.dev/skills/leolin990405/r-analytics-skill/h2o"><img src="https://agentmods.dev/badge/skills/leolin990405/r-analytics-skill/h2o/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/leolin990405/r-analytics-skill/h2o"><img src="https://agentmods.dev/badge/skills/leolin990405/r-analytics-skill/h2o.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 | $0.00025 | $0.00502 |
| Opus 5 | $0.00013 | $0.00251 |
| Sonnet 5 | $0.00005 | $0.00100 |
| Haiku 4.5 | $0.00003 | $0.00050 |
Grade A, and why
h2o 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 9d 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.
What it actually says
h2o Package
Scalable machine learning platform.
Initialize
library(h2o)
h2o.init(nthreads = -1, max_mem_size = "8G")
# Import data
df_h2o <- as.h2o(df)
df_h2o <- h2o.importFile("data.csv")
# Split
splits <- h2o.splitFrame(df_h2o, ratios = c(0.8), seed = 123)
train <- splits[[1]]
test <- splits[[2]]
AutoML
aml <- h2o.automl(
x = predictors,
y = "target",
training_frame = train,
max_runtime_secs = 300,
seed = 123
)
# Leaderboard
aml@leaderboard
# Best model
best <- aml@leader
h2o.performance(best, test)
Individual Models
# GLM
glm <- h2o.glm(x = predictors, y = "target",
training_frame = train, family = "binomial")
# Random Forest
rf <- h2o.randomForest(x = predictors, y = "target",
training_frame = train, ntrees = 100)
# GBM
gbm <- h2o.gbm(x = predictors, y = "target",
training_frame = train, ntrees = 100, learn_rate = 0.1)
# XGBoost
xgb <- h2o.xgboost(x = predictors, y = "target",
training_frame = train, ntrees = 100)
# Deep Learning
dl <- h2o.deeplearning(x = predictors, y = "target",
training_frame = train, hidden = c(200, 200))
Predictions
pred <- h2o.predict(model, test)
perf <- h2o.performance(model, test)
h2o.auc(perf)
h2o.confusionMatrix(perf)
Save/Load
h2o.saveModel(model, path = "models/")
model <- h2o.loadModel("models/model_id")
Shutdown
h2o.shutdown(prompt = FALSE)
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.
- 9d ago First seen · 90 lines · 25 tokens per session scan A 1817c054eae9
h2o is a skill published in the GitHub repository LeoLin990405/r-analytics-skill (5 stars, last pushed 6mo ago), licensed MIT. It adds 25 tokens to every session and 502 once invoked, about $0.0001 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 2026-09-03.
Other skills, from other repositories
ai-science-diffusion-generative-models
Code DDPM/DDIM diffusion samplers, linear/cosine noise schedules, and DDRM inverse-problem solving (denoising, inpainting, super-resolution) in NumPy/PyTorch. Use for forward/reverse diffusion, score matching, or DDIM sampling.
ai-science-esm2-embeddings
Generate ESM2 protein embeddings (fair-esm/transformers) and predict structure with ESMFold. Use when embedding sequences, scoring mutations zero-shot, annotating protein function, or doing fast MSA-free structure prediction.
ai-science-geneformer-scgpt
Tokenize scRNA-seq via Geneformer gene-rank or scGPT expression-bin encoding; annotate cell types, simulate in-silico knockouts. Use for foundation-model cell annotation, Geneformer/scGPT tokenization, or perturbation prediction.
ai-science-zero-shot-mutation
Score protein point mutations zero-shot with ESM-1v/ESM-2 masked-LM log-odds, ensembled, benchmarked on ProteinGym DMS. Use when predicting mutation effects, ranking missense variants, scoring VUS fitness with no labels.
bio-applied-cancer-transcriptomics
Classify tumor RNA-seq into subtypes (melanoma Tirosh/Harbst on TCGA-SKCM): log1p/z-score, PCA/t-SNE, hierarchical clustering, random forest, Kaplan-Meier survival. Use when subtyping cBioPortal expression data.
bio-applied-data-harmonization
Harmonize multi-omics data (RNA-seq, proteomics, methylation, metabolomics) before integration — per-layer normalization, KNN/half-minimum missing-value imputation, PCA/PVCA batch-effect detection, and ComBat correction with pandas/scikit-learn. Use when prepping matrices for MOFA2/DIABLO/mixOmics, fixing missing…