io.github.aion-org/edgar-answers-mcp

io.github.aion-org/edgar-answers-mcp

Provides normalized SEC EDGAR data (financials, filings, insider trades) to AI agents via MCP.

Category
访问服务器

README

<!-- mcp-name: io.github.aion-autonomous-org/edgar-answers-mcp -->

EDGAR Answers — SEC filings for AI agents

Give your agent answers from SEC EDGAR, not raw API responses.

This repository holds the documentation and registry manifest for a hosted MCP server. There is nothing to install or run: connect to the endpoint below from any MCP client. The server itself is operated as a managed service.

Every US public company files its financials with the SEC, for free, in XBRL. The catch is that "free and public" is not the same as "usable": companies tag the same line item a dozen different ways, annual and quarterly figures live in the same stream, restatements silently overwrite history, and Form 4 insider filings arrive as raw XML. Existing EDGAR MCP servers hand your agent that mess and wish it luck.

This server does the parsing first. Your agent asks for revenue; it gets revenue.

Tools

get_financials — normalized annual statements

Multi-year income statement, balance sheet, and cash flow for any US-listed company, from a ticker or CIK.

get_financials(company: "AAPL", years: 4)
→ revenue, grossProfit, operatingIncome, netIncome, epsDiluted,
  operatingCashFlow, capex, totalAssets, totalLiabilities,
  stockholdersEquity, cashAndEquivalents, longTermDebt, sharesDiluted
  — one clean row per fiscal year

Behind that: an ordered tag-fallback chain per line item (filers disagree on which us-gaap tag means "revenue"), duration filtering so annual figures aren't contaminated by quarterly cumulatives, and restatement resolution to the most recently filed value. Unreported items come back null rather than guessed.

filing_diff — what changed year over year

filing_diff(company: "NVDA", section: "risk_factors")
→ paragraphs added and removed between the two most recent 10-Ks

The question an analyst actually asks — what did they start saying this year, and what did they quietly drop? — answered at paragraph level, with boilerplate that merely reflowed filtered out.

insider_activity — Form 4, parsed

insider_activity(company: "MSFT", filings: 10)
→ who, role, transaction type, shares, price, holdings after — tidy rows

Open-market buys separated from tax withholding and option exercises, so your agent doesn't mistake a scheduled vest for a conviction purchase.

filing_section — any section as clean text

filing_section(company: "TSLA", section: "risk_factors" | "mdna" | "business" | "1A")

Table-of-contents decoys are filtered out; you get the real section body.

search_filings — full-text search since 2001

search_filings(query: "\"supply chain disruption\"", forms: "10-K", startDate: "2025-01-01")

Connect

Streamable HTTP endpoint:

https://aion-org--edgar-answers-mcp.apify.actor/mcp

Authenticate with your Apify API token as a bearer token. In an MCP client config:

{
  "mcpServers": {
    "edgar-answers": {
      "url": "https://aion-org--edgar-answers-mcp.apify.actor/mcp",
      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
    }
  }
}

For clients that only speak stdio, bridge with npx mcp-remote <url> --header "Authorization: Bearer YOUR_APIFY_TOKEN".

The server runs in standby mode: it scales to zero when idle and wakes on your first request, so the first call after a quiet period takes a few seconds.

Pricing

Pay per successful call — failed calls are not billed.

Call Price
search_filings $0.005
get_financials $0.02
insider_activity $0.02
filing_section $0.03
filing_diff $0.05

Data and compliance

Source data is SEC EDGAR — US government filings, public domain. Requests are made with a declared User-Agent and rate-limited well under the SEC's published ceiling, per their access guidelines. Nothing here is scraped from a site that prohibits it.

Limits, stated plainly

  • US GAAP filers. Foreign private issuers reporting under IFRS return no normalized financials; get_financials says so rather than guessing.
  • Coverage follows EDGAR. Full-text search reaches back to 2001; XBRL financials exist from roughly 2009 onward.
  • Not investment advice. This is filing data, faithfully parsed. What you conclude from it is yours.
  • filing_diff compares the two most recent filings of a form; arbitrary filing pairs aren't exposed yet.

Found a parsing bug or want a tool that isn't here? Open an issue here or on the Actor — specific reports about specific tickers get fixed.


Listed on the Apify Store. Documentation in this repository is © Aion; the service is proprietary.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选