MCP Server

Active

MIT

EdgarTools MCP Server

io.github.dgunning/edgartools

Free, open-source Python library for parsing SEC EDGAR filings as structured data — 20+ form types, no API key required.

Repository Registry

What is the EdgarTools MCP server?

EdgarTools is a Python library for accessing SEC EDGAR filings as structured data. It parses financial statements, insider trades, fund holdings, proxy statements, and 20+ other filing types into typed Python objects and pandas DataFrames. Built with an MCP server for AI integration, it requires no API key and provides free access to the complete SEC corpus since 1994.

EdgarTools turns unstructured SEC filings into ready-to-use Python objects. Extract income statements, balance sheets, insider transactions, 13F holdings, 8-K events, and more in a few lines of code. It's built for production pipelines, AI agents, and financial analysis — with smart caching, rate-limit awareness, and LLM-ready text output.

How to install EdgarTools

Copy-paste configuration for popular MCP clients.

transport: stdio

Config generated by PluginBench — verify against the source before use.

Environment / auth

EDGAR_IDENTITY
required
Your name and email for SEC EDGAR API identification (e.g. 'Your Name your.email@example.com')

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "edgartools": {
      "command": "uvx",
      "args": [
        "edgartools"
      ],
      "env": {
        "EDGAR_IDENTITY": "<YOUR_EDGAR_IDENTITY>"
      }
    }
  }
}

Cursor

~/.cursor/mcp.json

{
  "mcpServers": {
    "edgartools": {
      "command": "uvx",
      "args": [
        "edgartools"
      ],
      "env": {
        "EDGAR_IDENTITY": "<YOUR_EDGAR_IDENTITY>"
      }
    }
  }
}

Windsurf

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "edgartools": {
      "command": "uvx",
      "args": [
        "edgartools"
      ],
      "env": {
        "EDGAR_IDENTITY": "<YOUR_EDGAR_IDENTITY>"
      }
    }
  }
}

VS Code

.vscode/mcp.json

{
  "servers": {
    "edgartools": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "edgartools"
      ],
      "env": {
        "EDGAR_IDENTITY": "<YOUR_EDGAR_IDENTITY>"
      }
    }
  }
}

Claude Code

claude mcp add edgartools --env EDGAR_IDENTITY=<YOUR_EDGAR_IDENTITY> -- uvx edgartools

Tools & capabilities

Tools this server exposes to the agent.

get_financials — Extract standardized income statements, balance sheets, and cash flow statements from 10-K and 10-Q filings using XBRL data.
get_filings — Retrieve and parse any SEC filing type (Form 4, 13F, 8-K, DEF 14A, etc.) as typed objects with full history since 1994.
get_facts — Query XBRL financial data across companies by concept (e.g., Revenue) to build time-series DataFrames for cross-company comparison.
Company lookup — Retrieve company filings by ticker or CIK with industry and exchange filtering.
Insider trading extraction — Parse Form 3/4/5 filings to extract insider buy/sell transactions as structured data.
13F holdings parser — Extract institutional fund holdings and portfolio positions from 13F-HR filings.
8-K event extraction — Parse current reports to identify and structure reported corporate events and items.
Fund report parsing — Extract data from N-CSR, N-CEN, N-PORT, and N-MFP fund filings.
Section extraction — Extract specific sections like Risk Factors, MD&A, subsidiaries (EX-21), and auditor information from filings.
Text conversion — Convert HTML filings to clean text and markdown for RAG and full-text search.
MCP server — Built-in Model Context Protocol server for AI agents (Claude, Cline) to query SEC data in natural language.

Use cases

Compare revenue and profitability trends across companies using standardized XBRL financial statements
Track insider stock transactions and identify executive buying/selling patterns from Form 4 filings
Analyze institutional fund portfolios and holdings changes using 13F data
Monitor corporate events and material announcements through 8-K current reports
Build financial data pipelines and warehouses with typed objects and pandas DataFrames
Query SEC data via natural language through Claude or other AI agents using the built-in MCP server

EdgarTools MCP server FAQ

What is EdgarTools?

EdgarTools is a free, open-source Python library that parses SEC EDGAR filings into typed Python objects and pandas DataFrames. It supports 20+ filing types including 10-K, 10-Q, Form 4, 13F, 8-K, and proxy statements, with XBRL-standardized financial data for cross-company comparison.

Do I need an API key or subscription?

No. EdgarTools is completely free and requires no API key. You only need to set your email address once with `set_identity()` to identify yourself to the SEC, as required by EDGAR. There are no rate limits, quotas, or subscription tiers.

How do I install EdgarTools in Claude or Cursor?

Install via pip: `pip install "edgartools[ai]"`. For Claude Desktop, add the MCP server to your config file (`~/Library/Application Support/Claude/claude_desktop_config.json`) with the command `uvx --from edgartools[ai] edgartools-mcp`. For Claude Code, run `python -c "from edgar.ai import install_skill; install_skill()"` to add SEC analysis skills.

What filing types does EdgarTools support?

EdgarTools supports 20+ filing types including 10-K (annual), 10-Q (quarterly), 8-K (current events), Form 4 (insider trades), 13F (institutional holdings), N-CSR/N-CEN/N-PORT/N-MFP (fund reports), DEF 14A (proxy), S-1 (registration), Schedule 13D/G (ownership), Form D (offerings), Form C (crowdfunding), and Form 144 (restricted stock).

Can I use EdgarTools in production?

Yes. EdgarTools is production-ready with 1000+ tests, type hints throughout, smart caching, configurable rate limiting, and enterprise/academic mirror support. It's MIT-licensed and runs at hedge funds, fintechs, and research desks. Corporate sponsorship tiers are available for teams needing SLAs and support.

How does EdgarTools compare to other SEC data services?

Unlike hosted APIs like sec-api (which charge $49+/mo), EdgarTools is a free, open-source library you run locally. It provides typed Python objects instead of raw JSON, includes built-in AI/MCP support, and gives you full control with no vendor lock-in or API keys.

README (reference)

Source of truth, from the repository.

EdgarTools — Python Library for SEC EDGAR Filings

EdgarTools is a Python library for accessing SEC EDGAR filings as structured data. Parse financial statements, insider trades, fund holdings, proxy statements, and 20+ other filing types with a consistent Python API — in a few lines of code. Free and open source.

EdgarTools SEC filing data extraction demo

Why EdgarTools?

SEC EDGAR has every filing back to 1994, free — and almost none of it is ready to use. EdgarTools turns any filing into a typed Python object, so a 10-K's revenue is one line instead of an afternoon of XBRL parsing.

# Apple's latest income statement — rendered, standardized, done
from edgar import Company
Company("AAPL").get_financials().income_statement()

<table align="center"> <tr> <td align="center" width="33%"> <img src="https://raw.githubusercontent.com/dgunning/edgartools/main/docs/images/icons/icon-data.svg" width="96" alt="Financial Statements"> Financial Statements Income, balance sheet, cash flow in one call XBRL-standardized for cross-company comparison </td> <td align="center" width="33%"> <img src="https://raw.githubusercontent.com/dgunning/edgartools/main/docs/images/icons/icon-filings.svg" width="96" alt="Every Filing Type"> Every Filing Type 13F holdings, Form 4 insiders, 8-K events, funds, proxies Typed objects + pandas DataFrames for 20+ forms </td> <td align="center" width="33%"> <img src="https://raw.githubusercontent.com/dgunning/edgartools/main/docs/images/icons/icon-ai.svg" width="96" alt="Built for Pipelines & AI"> Built for Pipelines & AI Rate-limit aware, smart caching, enterprise mirrors Built-in MCP server + LLM-ready text for RAG </td> </tr> </table>

How It Works

Everything starts with a Company or a Filing. Call .obj() and you get a typed object built for that form — its data ready as pandas DataFrames and clean text.

The same typed output that reads cleanly in a notebook drops straight into a pipeline: DataFrames for your warehouse, LLM-ready text and an MCP server for your AI stack, rate-limit and enterprise-mirror aware for scale.

Quick Start

1. Install

pip install edgartools

2. Identify yourself to the SEC — EDGAR requires an email with every request. No key, no signup, no rate-limit tier; set it once:

from edgar import *
set_identity("your.name@example.com")

3. Get data — every filing is now a few lines away:

# Standardized financial statements, straight from XBRL
Company("AAPL").get_financials().income_statement()

# The latest insider Form 4 as a structured object
Company("AAPL").get_filings(form="4").latest().obj()

Apple SEC Form 4 insider transactions parsed into a structured Python object

Next: explore the Use Cases below, or dive into the documentation and Quick Guide.

Use Cases

Financial statements from 10-K and 10-Q filings

financials = Company("MSFT").get_financials()
financials.balance_sheet()     # all line items
financials.income_statement()  # revenue, net income, EPS

Financial Statements guide →

Insider trading from SEC Form 4

form4 = Company("TSLA").get_filings(form="4").latest().obj()
form4.to_dataframe()  # insider buy/sell transactions

Insider Trades guide →

13F institutional holdings & hedge fund portfolios

thirteenf = get_filings(form="13F-HR").latest().obj()
thirteenf.holdings  # every portfolio position as a DataFrame

Institutional Holdings guide →

8-K current reports & corporate events

eightk = get_filings(form="8-K").latest().obj()
eightk.items  # reported event items

Current Events guide →

XBRL financial data across companies

facts = Company("AAPL").get_facts()
facts.query().by_concept("Revenue").to_dataframe()  # revenue history as a DataFrame

XBRL Deep Dive →

Key Features

Financial data

Income, balance sheet, cash flow — XBRL-standardized for cross-company comparison
Individual line items, dimensional data, multi-period comparatives
Company Facts API: time-series for any concept across years

Funds & ownership

13F holdings, N-PORT, N-MFP, N-CSR/N-CEN fund reports
Form 3/4/5 insider transactions; Schedule 13D/G ownership
Position tracking over time

</td> <td width="50%" valign="top">

Filings & text

Typed objects for 20+ forms; complete history since 1994
Section extraction (Risk Factors, MD&A), EX-21 subsidiaries, auditor info
HTML → clean text + markdown for RAG; full-text search
Ticker/CIK lookup, industry & exchange filtering

Built for production

Configurable rate limiting + enterprise/academic mirrors
Smart caching, type hints throughout, 1000+ tests
Enterprise configuration →

</td> </tr> </table>

EdgarTools supports all SEC form types including 10-K annual reports, 10-Q quarterly filings, 8-K current reports, 13F institutional holdings, Form 4 insider transactions, proxy statements (DEF 14A), S-1 registration statements, N-CSR fund reports, N-MFP money market data, N-PORT fund portfolios, Schedule 13D/G ownership, Form D offerings, Form C crowdfunding, and Form 144 restricted stock. Parse XBRL financial data, extract text sections, and convert filings to pandas DataFrames.

Comparison with Alternatives

EdgarTools is a Python library that talks directly to SEC EDGAR. sec-api is the best-known hosted API that returns JSON. Both parse filings — the difference is how you work with the data, and what it costs you.

	EdgarTools	sec-api
Cost	Free, MIT	$49+/mo
Data format	Typed Python objects → DataFrames	JSON you parse yourself
Where it runs	In your process — no key, no quotas, no vendor lock-in	Hosted API — key + rate tiers
Filing coverage	20+ typed forms (10-K, 8-K, 13F, N-PORT, proxy…)	15+ structured endpoints
AI / MCP	<img src="https://raw.githubusercontent.com/dgunning/edgartools/main/docs/images/icons/compare-check.svg" width="20"> Built in	<img src="https://raw.githubusercontent.com/dgunning/edgartools/main/docs/images/icons/compare-cross.svg" width="20">
Open source	<img src="https://raw.githubusercontent.com/dgunning/edgartools/main/docs/images/icons/compare-check.svg" width="20"> Inspect, fork, self-host	<img src="https://raw.githubusercontent.com/dgunning/edgartools/main/docs/images/icons/compare-cross.svg" width="20"> Proprietary

Bottom line: in Python, EdgarTools gives you typed objects, AI-native output, and the full SEC corpus — free, open, and inspectable, with no keys or bills. pip install edgartools and you're querying filings in two lines.

Library or hosted?

EdgarTools is the open-source library — SEC-filing primitives you compose in your own code, free and self-run.

edgar.tools is the hosted platform built on that same open engine: the full SEC corpus as a managed service, so your team gets the data without running the pipeline — and without the black box of a closed API.

Reach for the library when you want control in your own stack; reach for edgar.tools when you'd rather not operate it yourself.

AI Integration

Use EdgarTools with Claude Code & Claude Desktop

EdgarTools includes an MCP server and AI skills for Claude Desktop and Claude Code. Ask questions in natural language and get answers backed by real SEC data.

"Compare Apple and Microsoft's revenue growth rates over the past 3 years"
"Which Tesla executives sold more than $1 million in stock in the past 6 months?"

<details> <summary>Setup Instructions</summary>

Option 1: AI Skills (Recommended)

Install the EdgarTools skill for Claude Code or Claude Desktop:

pip install "edgartools[ai]"
python -c "from edgar.ai import install_skill; install_skill()"

This adds SEC analysis capabilities to Claude, including 3,450+ lines of API documentation, code examples, and form type reference.

Option 2: MCP Server

Run EdgarTools as an MCP server for any AI client -- Claude Desktop, Cline, or your own containerized deployment.

Add to Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "edgartools": {
      "command": "uvx",
      "args": ["--from", "edgartools[ai]", "edgartools-mcp"],
      "env": {
        "EDGAR_IDENTITY": "Your Name your.email@example.com"
      }
    }
  }
}

Requires uv. Alternatively, pip install "edgartools[ai]" and use python -m edgar.ai.

See AI Integration Guide for complete documentation.

</details>

❤️ Support This Project

EdgarTools runs in production at hedge funds, fintechs, and research desks — MIT-licensed, no keys, no subscriptions, and maintained by one person.

The SEC amends filing formats every quarter and ships a new XBRL taxonomy every year. Sponsorship is what keeps 20+ parsers current and funds new extractors as fresh disclosure types appear.

<a href="https://github.com/sponsors/dgunning" target="_blank"> <img src="https://img.shields.io/badge/Sponsor-30363D?style=for-the-badge&logo=GitHub-Sponsors&logoColor=EA4AAA" alt="Sponsor on GitHub" height="44"> </a>    <a href="https://www.buymeacoffee.com/edgartools" target="_blank"> <img src="https://img.shields.io/badge/Buy_me_a_coffee-FFDD00?style=for-the-badge&logo=buymeacoffee&logoColor=black" alt="Buy Me A Coffee" height="44"> </a> Recurring sponsorship + corporate tiers via GitHub · One-time thanks via Buy Me a Coffee

For teams running EdgarTools in production

If EdgarTools is in your data pipeline, GitHub Sponsors offers corporate tiers from $250 to $1,500/mo with:

Response SLAs (24h–48h first response on critical issues)
Quarterly strategy calls and roadmap input
Logo placement in this README
7-day early access for internal regression testing
Annual invoicing through GitHub — procurement-friendly

→ See sponsor tiers

Community & Support

Documentation & Resources

Get Help & Connect

GitHub Issues - Bug reports and feature requests
Discussions - Questions and community discussions

Contributing

Contributions welcome:

Code: Fix bugs, add features, improve documentation
Examples: Share interesting use cases and examples
Feedback: Report issues or suggest improvements
Spread the Word: Star the repo, share with colleagues

See our Contributing Guide for details.

Professional Services

Need help building production SEC data infrastructure? The creator of EdgarTools offers consulting for teams building financial AI products:

SEC Data Sprint (1–3 days) — Working prototype on your data
Architecture Review (1–2 weeks) — Pipeline audit with prioritized fixes
Pipeline Build (2–4 weeks) — Production-ready code, tests, and handoff

Learn more →

EdgarTools is distributed under the <a href="LICENSE">MIT License</a>

Star History

View repository →