History of Market
Read-only U.S. stock market history: S&P 500 since 1928, CAPE, drawdowns, index changes.
我該用這個嗎
品質與安全性
發現項目（1）
- LOWTool 'annual_returns' description lacks action verb在 annual_returns 中
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高，可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案：
{
"mcpServers": {
"market-history": {
"url": "https://historyofmarket.com/mcp"
}
}
}遠端端點
https://historyofmarket.com/mcpstreamable-http它能做什麼
工具清單
工具（5）
🟢get_latest(id)
Current headline readings (value, unit, observation date) for the site's charts: S&P 500 level, Shiller CAPE, forward P/E for the S&P 500, Nasdaq 100, SOX and sector ETFs, VIX, drawdown from the high, breadth, Magnificent 7 weight and more. Pass `id` for one chart, or omit it for all.
輸入結構描述
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Chart id, e.g. \"pe\" (Shiller CAPE), \"forward-pe\", \"ndx-forward-pe\", \"drawdown\", \"vix\". Omit for every reading."
}
},
"additionalProperties": false
}🟢annual_returns(index, from_year, to_year)
Calendar-year returns by year, price return and total return (dividends reinvested), with the arithmetic mean and compound annual rate over the complete years in the range. Indexes: sp500 (from 1928), ndx, sox, xlk, xlf, xly, xlp. The current year is year-to-date and flagged `ytd`.
輸入結構描述
{
"type": "object",
"properties": {
"index": {
"type": "string",
"enum": [
"sp500",
"ndx",
"sox",
"xlk",
"xlf",
"xly",
"xlp"
],
"description": "Which index or ETF. Default sp500."
},
"from_year": {
"type": "integer",
"description": "First year to include."
},
"to_year": {
"type": "integer",
"description": "Last year to include."
}
},
"additionalProperties": false
}🟢index_changes(index, since, ticker, limit)
Constituent changes with effective date, the company added, the company removed and the reason. sp500 from 2016, ndx from 2020. Newest first.
輸入結構描述
{
"type": "object",
"properties": {
"index": {
"type": "string",
"enum": [
"sp500",
"ndx"
],
"description": "Default sp500."
},
"since": {
"type": "string",
"description": "Only changes effective on or after this date (YYYY-MM-DD)."
},
"ticker": {
"type": "string",
"description": "Only changes that add or remove this ticker."
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 300,
"description": "Default 20."
}
},
"additionalProperties": false
}🟢search_charts(query, limit)
Search the site's charts by topic or question (e.g. "bear markets", "Shiller CAPE", "Nasdaq 100 drawdown", "monthly seasonality"). Returns each chart's title, the question it answers, its page and the dataset path to pass to get_dataset.
輸入結構描述
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Words to match against chart titles and the questions they answer."
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 50,
"description": "Default 10."
}
},
"required": [
"query"
],
"additionalProperties": false
}🟢get_dataset(path, since, until, tail)
Read one published dataset (the JSON behind a chart), e.g. "sp500/drawdowns", "sp500/pe", "ndx/forward-pe", "recessions". Rows can be filtered by `since`/`until` on their `date` (or `year`). A result is capped at about 60,000 characters: when a file is larger (long daily series), each array keeps its most recent rows and the cut is reported under `windowed`. Scalars and summary fields are always included.
輸入結構描述
{
"type": "object",
"properties": {
"path": {
"type": "string",
"description": "Dataset path without /api/ and .json, as returned by search_charts."
},
"since": {
"type": "string",
"description": "Keep rows dated on or after this (YYYY-MM-DD or YYYY)."
},
"until": {
"type": "string",
"description": "Keep rows dated on or before this (YYYY-MM-DD or YYYY)."
},
"tail": {
"type": "integer",
"minimum": 1,
"maximum": 2000,
"description": "Rows kept per array after filtering, from the end. By default every row is returned when the result fits; long daily series are cut to their most recent rows and reported under `windowed`."
}
},
"required": [
"path"
],
"additionalProperties": false
}社群
證據