Context7
Up-to-date docs for AI agents
作者提供的说明(英文原文)
SEC EDGAR filings (10-K, 10-Q, 8-K), XBRL financials, and full-text filing search. No API key.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
SEC_USER_AGENT你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"edgar": {
"command": "npx",
"args": [
"-y",
"edgar-mcp"
],
"env": {
"SEC_USER_AGENT": "your-value"
}
}
}
}在终端中运行这条命令
claude mcp add edgar --env SEC_USER_AGENT=your-value -- npx -y edgar-mcp运行前请先填写空着的值。
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"edgar": {
"command": "npx",
"args": [
"-y",
"edgar-mcp"
],
"env": {
"SEC_USER_AGENT": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"edgar": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"edgar-mcp"
],
"env": {
"SEC_USER_AGENT": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
SEC_USER_AGENT env这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日
注册名称已通过 GitHub 账户 danishashko 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于–,最新版本(1.0.2)于5个月前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
同一分类中有:Context7, Codebase Memory, Exa
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Up-to-date docs for AI agents
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Connect AI agents to Exa for web search, content fetching, and multi-step research.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
U.S. SEC filings and financial data for Claude Desktop and any MCP-compatible client, powered by SEC EDGAR. Search companies, pull filings (10-K, 10-Q, 8-K, Form 4, ...), read XBRL financials, and full-text search across all filings — all from natural language. No API key required.
npm package:
edgar-mcp· GitHub repo:danishashko/edgar-mcp.
Every tool returns human-readable markdown by default, or structured JSON on request (response_format: "json"). Lightweight (Python standard library + mcp only), no API key, retries on rate limits.
Add this to your Claude Desktop config and restart Claude:
~/Library/Application Support/Claude/claude_desktop_config.json%APPDATA%\Claude\claude_desktop_config.json{
"mcpServers": { "edgar": { "command": "npx", "args": ["-y", "edgar-mcp"], "env": { "SEC_USER_AGENT": "your-app-name your@email.com" } } } }
The SEC asks every API client to send a descriptive User-Agent with contact info. Set SEC_USER_AGENT to your own name/email (a default is provided so it works out of the box). On first launch the npx wrapper creates an isolated Python environment and installs the dependency (one-time, ~a minute). You need Python 3.10+ and Node.js 16+.
npm install -g edgar-mcp
{
"mcpServers": { "edgar": { "command": "edgar-mcp", "env": { "SEC_USER_AGENT": "your-app-name your@email.com" } } } }
| Tool | What it returns | Parameters |
|---|---|---|
search_company | Companies matching a name/ticker, with their CIK | query, limit |
get_recent_filings | Recent filings (form, dates, link), optional form filter | identifier, form_type, limit |
get_latest_filing | The newest filing of a given type, with a document link | identifier, form_type |
get_company_facts | Headline financials (revenue, net income, assets, equity, cash, EPS) | identifier |
get_concept | One XBRL concept's reported history over time | identifier, concept, limit |
full_text_search | Filings matching a keyword/phrase across all companies | query, forms, limit |
Every tool also accepts response_format ("markdown", the default, or "json"). identifier accepts a ticker, CIK, or company name.
Common XBRL concepts for `get_concept`: Revenues / RevenueFromContractWithCustomerExcludingAssessedTax, NetIncomeLoss, Assets, Liabilities, StockholdersEquity, EarningsPerShareDiluted.
Once the server is connected, just ask Claude:
"SEC EDGAR is rate-limiting requests" EDGAR allows ~10 requests/second. The server retries automatically; wait a moment if it persists.
"Not found" Use search_company to confirm the exact ticker/CIK first.
"Command not found" / "Python not found" Ensure Python 3.10+ and Node.js 16+ are installed and on your PATH. On macOS/Linux, try python3.
Tools not showing up in Claude
Using a different model or provider (LiteLLM, OpenRouter, NVIDIA NIM, a local model) This server never talks to a model. Your client starts it as a local process and speaks JSON-RPC over stdin/stdout, so changing ANTHROPIC_BASE_URL or swapping the model behind your client has no effect on it. If tools stop firing after a switch like that, check two things: the model has to support function calling, and a proxy configured to drop unsupported parameters can silently strip your tool definitions, which produces no error at all. Run /mcp in your client (or claude mcp list) to confirm the server is connected before suspecting the server.
If you would rather run the Python file directly instead of via npx:
1. Download the server and install the dependency
pip install mcp
(or pip3 on macOS/Linux)
2. Point Claude Desktop at it
{
"mcpServers": { "edgar": { "command": "python3", "args": ["/absolute/path/to/edgar_mcp.py"], "env": { "SEC_USER_AGENT": "your-app-name your@email.com" } } } }
On Windows use "command": "python" and a path like "C:\\path\\to\\edgar_mcp.py".
3. Restart Claude Desktop.
User-Agent with contact info and limits ~10 requests/second.identifier is flexible: ticker (AAPL), CIK (320193), or name (Apple).get_company_facts returns highlights; use get_concept for the full series of any one metric.See CHANGELOG.md for the full version history.
This tool uses the public SEC EDGAR APIs but is not affiliated with or endorsed by the U.S. Securities and Exchange Commission. Use is subject to the SEC's website policies. Data is provided as-is for personal, educational, and research purposes; verify against the original filings before relying on it.
Daniel Shashko
MIT © Daniel Shashko