databutler-stats
Exact statistics & probability: distributions, hypothesis tests, CIs, Bayesian updates, regression.
使うべきか
品質と安全性
検出事項（1）
- LOWTool 'confidence_interval' description lacks action verbconfidence_interval 内
ツール定義とプロトコルへの準拠に関する自動分析に基づいています。
コンテキストコスト
これは、サーバーのツールがモデルのコンテキストに読み込まれるたびに消費されるおおよそのトークン数です。数が多いほど、ほかのタスクに使える注意が減ります。
インストール
ワンクリックインストール
これを `claude_desktop_config.json` ファイルに追加してください:
{
"mcpServers": {
"databutler-stats": {
"url": "https://databutler.dev/api/mcp/stats"
}
}
}リモートエンドポイント
https://databutler.dev/api/mcp/statsstreamable-httpできること
ツール一覧
ツール（6）
⚪descriptive_stats(data)
Summary statistics for a numeric array: mean, median, sd, variance, quartiles, IQR, skewness, min/max.
入力スキーマ
{
"type": "object",
"properties": {
"data": {
"type": "array",
"items": {
"type": "number"
}
}
},
"required": [
"data"
]
}⚪distribution(dist, params, at, p)
Evaluate a probability distribution (normal, t, chi2, binomial, poisson): pdf/pmf and cdf at a value, and/or the quantile at a probability, plus mean & variance. Params per dist: normal {mean,sd}, t {df}, chi2 {df}, binomial {n,p}, poisson {lambda}.
入力スキーマ
{
"type": "object",
"properties": {
"dist": {
"type": "string",
"enum": [
"normal",
"t",
"chi2",
"binomial",
"poisson"
]
},
"params": {
"type": "object"
},
"at": {
"type": "number",
"description": "value to evaluate pdf/pmf and cdf at"
},
"p": {
"type": "number",
"description": "probability to get the quantile for (0-1)"
}
},
"required": [
"dist"
]
}🟢hypothesis_test(test, data, data1, data2, mu0, ...)
Run a significance test and get the statistic, p-value, and a plain-language interpretation with assumptions. test = one-sample-t {data, mu0}, two-sample-t {data1, data2}, one-proportion-z {successes, n, p0}, two-proportion-z {successes1,n1,successes2,n2}, chi2-gof {observed, expected?}, chi2-independence {table}. Optional tail: two-sided (default) | greater | less; alpha default 0.05.
入力スキーマ
{
"type": "object",
"properties": {
"test": {
"type": "string"
},
"data": {
"type": "array",
"items": {
"type": "number"
}
},
"data1": {
"type": "array",
"items": {
"type": "number"
}
},
"data2": {
"type": "array",
"items": {
"type": "number"
}
},
"mu0": {
"type": "number"
},
"successes": {
"type": "number"
},
"n": {
"type": "number"
},
"p0": {
"type": "number"
},
"successes1": {
"type": "number"
},
"n1": {
"type": "number"
},
"successes2": {
"type": "number"
},
"n2": {
"type": "number"
},
"observed": {
"type": "array",
"items": {
"type": "number"
}
},
"expected": {
"type": "array",
"items": {
"type": "number"
}
},
"table": {
"type": "array"
},
"tail": {
"type": "string"
},
"alpha": {
"type": "number"
}
},
"required": [
"test"
]
}⚪confidence_interval(kind, confidence, data, n, mean, ...)
Confidence interval for a mean (t-based; from data, or n/mean/sd) or a proportion (Wilson; successes/n). kind = mean | proportion; confidence default 0.95.
入力スキーマ
{
"type": "object",
"properties": {
"kind": {
"type": "string",
"enum": [
"mean",
"proportion"
]
},
"confidence": {
"type": "number"
},
"data": {
"type": "array",
"items": {
"type": "number"
}
},
"n": {
"type": "number"
},
"mean": {
"type": "number"
},
"sd": {
"type": "number"
},
"successes": {
"type": "number"
}
}
}🟡bayes_update(hypotheses)
Discrete Bayesian update: given competing hypotheses each with a prior and the likelihood of the observed evidence, return normalised posteriors. Priors are renormalised to sum to 1.
入力スキーマ
{
"type": "object",
"properties": {
"hypotheses": {
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {
"type": "string"
},
"prior": {
"type": "number"
},
"likelihood": {
"type": "number"
}
},
"required": [
"prior",
"likelihood"
]
}
}
},
"required": [
"hypotheses"
]
}⚪linear_regression(x, y)
Simple linear regression of y on x: slope, intercept, r, r², slope std error and p-value, equation.
入力スキーマ
{
"type": "object",
"properties": {
"x": {
"type": "array",
"items": {
"type": "number"
}
},
"y": {
"type": "array",
"items": {
"type": "number"
}
}
},
"required": [
"x",
"y"
]
}コミュニティ
エビデンス