edgartools SEC MCP Server
Local MCP server for the edgartools Python library, letting Claude search, parse, read, compare, and analyze SEC EDGAR filings, companies, ownership, funds, proxy statements, notes, trends, and live filing feeds.
Open the source and read safety notes before installing.
Safety notes
- edgartools MCP can query live SEC filings, parse structured filing data, search filing text, monitor recent filings, compare companies, inspect ownership, and extract proxy or fund information.
- SEC EDGAR access requires an identity string and can be rate-limited if automated requests are too frequent or unidentified.
- Filing analysis can produce plausible financial, legal, governance, compensation, or risk conclusions that still require expert review.
- The local server is stateless, but remote HTTP transport deployments still need normal network access controls if shared with a team.
- Do not treat model-generated summaries as investment advice, legal advice, accounting advice, or a substitute for reading the filing source.
Privacy notes
- Tool inputs can reveal watchlists, portfolio interests, deal research, diligence targets, litigation topics, trading hypotheses, or compliance questions.
- Tool outputs can expose filing excerpts, insider transactions, institutional holdings, compensation data, financial metrics, and query strategy in the MCP client transcript.
- EDGAR_IDENTITY includes a name and email address; keep it scoped to approved automation and avoid committing it to source control.
- If using hosted edgar.tools services instead of the local server, review the hosted service's tier, retention, processing, and data-use terms separately.
Prerequisites
- Python 3.10 or newer and `uvx` available, or the `edgartools[ai]` package installed.
- EDGAR_IDENTITY set to the name and email required for automated SEC EDGAR access.
- Network access to SEC EDGAR endpoints and patience for rate limits or transient filing fetch failures.
- Understanding of the filing forms, companies, sections, and time periods being analyzed.
- Human review before using SEC filing analysis for investment, legal, accounting, compliance, or disclosure decisions.
Schema details
- Install type
- cli
- Troubleshooting
- No
- Scope
- Source repo
- Estimated setup
- 10 minutes
- Difficulty
- intermediate
Full copyable content
{
"mcpServers": {
"edgartools": {
"command": "uvx",
"args": ["--from", "edgartools[ai]", "edgartools-mcp"],
"env": {
"EDGAR_IDENTITY": "Your Name your.email@example.com"
}
}
}
}About this resource
Content
edgartools SEC MCP Server connects Claude to the edgartools Python library for local SEC EDGAR research. It can retrieve company profiles, search filing metadata and full text, monitor recent filings, parse specific filings, read sections, inspect notes, analyze financial trends, compare companies, review ownership, look at fund holdings, and examine proxy compensation or governance data.
The local MCP server is installed through the edgartools[ai] extra and runs
with stdio by default. The setup docs also describe Streamable HTTP transport
for shared deployments and distinguish the local open-source server from the
hosted edgar.tools MCP service.
Source Review
- https://github.com/dgunning/edgartools
- https://github.com/dgunning/edgartools/blob/main/docs/ai/mcp-setup.md
- https://pypi.org/project/edgartools/
- https://github.com/dgunning/edgartools/blob/main/docs/ai/mcp-tools.md
- https://github.com/dgunning/edgartools/blob/main/pyproject.toml
- https://github.com/dgunning/edgartools/blob/main/LICENSE.txt
These sources were reviewed on 2026-06-06. Prefer the live repository, MCP setup guide, PyPI page, MCP tools reference, package metadata, and license for current install commands, tool names, SEC identity requirements, and transport options.
Features
- Local stdio MCP server through
uvx --from "edgartools[ai]" edgartools-mcp. EDGAR_IDENTITYconfiguration for SEC automated access identification.- 13 documented tools organized around SEC research intent.
- Company profile, financial, filing, and ownership lookup through
edgar_company. - Company and filing search through
edgar_search. - Company screening by industry, exchange, state, or SIC through
edgar_screen. - Full-text filing search through
edgar_text_search. - Recent filing monitoring through
edgar_monitor. - Structured filing parsing and section extraction through
edgar_filingandedgar_read. - Notes, trends, company comparison, ownership, fund, and proxy tools.
- Optional Streamable HTTP transport for shared deployments.
Installation
For Claude Code:
claude mcp add edgartools -- uvx --from "edgartools[ai]" edgartools-mcp
Set the SEC identity required for automated EDGAR access:
export EDGAR_IDENTITY="Your Name your.email@example.com"
For MCP clients that use JSON configuration:
{
"mcpServers": {
"edgartools": {
"command": "uvx",
"args": ["--from", "edgartools[ai]", "edgartools-mcp"],
"env": {
"EDGAR_IDENTITY": "Your Name your.email@example.com"
}
}
}
}
Restart the MCP client after adding the server.
Use Cases
- Ask Claude to summarize a company's latest 10-K, 10-Q, or 8-K with filing identifiers.
- Search recent filings for cybersecurity, AI, restructuring, revenue recognition, or risk-factor language.
- Pull a specific filing section such as MD&A, risk factors, financials, controls, legal proceedings, or proxy compensation.
- Compare financial trends across companies or periods using XBRL-sourced data.
- Monitor recent SEC filings for a form type, company, or event-driven research workflow.
- Inspect insider transactions, institutional holdings, fund data, or proxy governance details.
Safety and Privacy
SEC filings are public, but research intent is not. Treat prompts and tool calls
as sensitive if they reveal portfolio targets, diligence questions, legal
theories, competitive monitoring, or trading hypotheses. Keep EDGAR_IDENTITY
out of committed files and use an approved automation identity.
Do not let a model's filing summary become the final decision. Verify citations, accession numbers, section text, filing dates, and financial figures against the source filing before using results for investment, legal, accounting, compliance, governance, or public reporting work.
Duplicate Check
Existing financial MCP entries cover hosted financial datasets, standardized
market data, or curated provider databases such as Financial Datasets, Drillr,
and Daloopa. This entry covers the separate local MIT-licensed edgartools Python
library and MCP server for SEC EDGAR filings. No dgunning/edgartools,
edgartools-mcp, or dedicated edgartools SEC MCP entry was found in
content/mcp, content/tools, content/guides, content/agents, or
content/skills.
Source citations
Signals
Loading live community signals…
A short, calm digest of reviewed Claude resources. Unsubscribe any time.