SkillFoxx

EdgarTools MCP

A Python MCP server for SEC EDGAR: 10-K and 8-K filings, XBRL financials, insider trades and fund portfolios, no API key.

  • 2,846 ★ on the repository
  • Last commit
MCP server

Minimal access

Why this level

  • Reads public SEC data and writes nothing; only the identity string leaves the machine.
All reasons and checks

dgunning/edgartoolsAuthor:dgunning

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add mcp/edgartools

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Minimal accessThe tool runs nothing and makes no network calls.

  • Reads public SEC data and writes nothing; only the identity string leaves the machine.

This is third-party code. Review the repository files before installing.

What it does

EdgarTools reads SEC EDGAR data and parses US company documents into structured form. The Python library extracts financial statements, Forms 3, 4 and 5, 13F portfolios, ADV documents and sections of annual reports. The `ai` extra adds an MCP server with thirteen tools and seven ready prompts. Tools are grouped by task: `edgar_company` gathers a profile, financials, filings and ownership in one call, `edgar_trends` builds metric series with growth rates, `edgar_compare` compares companies, `edgar_text_search` searches document text, and `edgar_ownership` shows insider trades and institutional portfolios. Running it needs only an identity string for the SEC, and no API key.

Who it is for. Analysts and independent researchers who go through US company filings together with an agent and do not read documents by hand on the SEC site.

Good fit when

  • You need to compare revenue and margin of several issuers over four quarters.
  • An agent summarises the risk factors section of a fresh 10-K.
  • An analyst follows insider trades on a ticker through Form 4 filings.
  • A researcher looks at which positions a fund disclosed in its 13F report.

Not a fit when

  • You need real-time quotes and trading data: EDGAR serves disclosures only.
  • You need filings of non-US companies.
  • An investment decision would be made without checking the figures against the source document.

Example request

Compare Apple and Microsoft revenue and operating margin over the last four quarters and show where research spending grew.

MCP

Transport
stdio
Authentication
not required

Environment variables

EDGAR_IDENTITY required
Name and email that identify the client to SEC EDGAR.

Limitations

The library needs Python 3.11 or newer, and version 6.0 renamed some 5.x methods. Before starting set `EDGAR_IDENTITY` in the format "Name email": the SEC requires client identification. Claude Desktop needs the full path to `python3` or `uvx` because the app does not inherit the shell PATH. Data comes from the SEC site and follows its request rate limits. The results are not investment advice.

How to disable. Remove the edgartools entry from the MCP configuration (`claude mcp remove edgartools`). Uninstall the package with `pip uninstall edgartools` and drop `EDGAR_IDENTITY` from the environment.

FAQ

How do I connect EdgarTools to Claude Code?

Run `claude mcp add edgartools -- uvx --from "edgartools[ai]" edgartools-mcp` and export `EDGAR_IDENTITY`. For Claude Desktop add the same launch to `claude_desktop_config.json` with the full path to `uvx`.

Is an API key required?

No. An `EDGAR_IDENTITY` string with your name and email is enough, since the SEC uses it to identify the client.

What does `edgar_company` do?

It is the entry point for company questions: one call returns a profile, financials, recent filings and ownership. The `include` parameter picks sections and `period` sets annual, quarterly or trailing twelve months.

How does it differ from the hosted edgar.tools connector?

Here the server runs locally on your machine, while the edgar.tools connector is hosted at app.edgar.tools. Attach one of the two, since they overlap and confuse tool routing.

Is there a skill alongside the MCP server?

Yes. The command `python -c "from edgar.ai import install_skill; install_skill()"` installs a skill for Claude Code and Claude Desktop. Both the skill and the MCP server ship in the `ai` extra.

Discussion

No questions or reviews yet.

MCP servers

An MCP server for SEC EDGAR: US company filings, forms 10-K, 10-Q and 8-K, XBRL-parsed financials and insider trades

MCP serverBroad access387

MCP server over Korean DART and US SEC EDGAR disclosures: company financials, peer comparison and valuation for an agent.

MCP serverExtended access227
Official

A financial analysis MCP server: 500+ methods from company statements, ratios, indicators, risk metrics and macro data

MCP serverExtended access5.5K
Official

RoboSystems graph MCP endpoint: an agent queries company financial data and SEC XBRL filings in a knowledge graph

MCP serverBroad access30

Financial Datasets MCP

Financial Datasets MCP Server

Official

An MCP server for the Financial Datasets API: company financials, stock and crypto prices and news for the agent

MCP serverExtended access2.3K

An MCP server that screens companies on metrics from filings as of any past date, with no look-ahead

MCP serverExtended access331

README in short

The library parses SEC forms into typed objects and is MIT licensed. The `ai` extra adds a local MCP server and a skill for Claude Code and Claude Desktop.

README badge

Report a problem