SEC EDGAR MCP
Query SEC EDGAR filings — 10-K/10-Q/8-K sections, XBRL financials, insider trades — with exact numbers and source URLs.
Add to your client
Copy the config for your MCP client and paste it into its config file.
docker run -i --rm -e "SEC_EDGAR_USER_AGENT=Your Name (your@email.com)" stefanoamorelli/sec-edgar-mcp:latestPaste into ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"sec-edgar-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"SEC_EDGAR_USER_AGENT=Your Name (your@email.com)",
"stefanoamorelli/sec-edgar-mcp:latest"
]
}
}
}Requires Docker to be installed and running.
Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf
Before you start
- Docker (quick start) or Python 3.11+ for pip/conda/uv installs
- A SEC_EDGAR_USER_AGENT string with your name and email (SEC fair-access requirement, not a key)
- No account or API key — EDGAR is free public data
About SEC EDGAR MCP
SEC EDGAR MCP wraps the SEC's public EDGAR system for AI agents, with a design goal the README states up front: exact numeric precision. Financial statements are parsed from XBRL rather than scraped text, so an agent quoting Apple's revenue gets the filed figure, and every tool response carries the underlying SEC URL for verification — a sensible guard against hallucinated financials.
The tool surface groups into four areas: company tools (CIK lookup, company info, company facts), filing tools (10-K/10-Q/8-K retrieval plus section extraction), financial tools (balance sheet, income statement, cash flow), and insider tools (Form 3/4/5 transactions). That covers the workhorse questions of fundamental analysis: what did the company file, what do the statements say, and who bought or sold stock.
Transport is stdio by default via Docker (the -i flag is required for JSON-RPC over stdin). For platforms like Dify there is a streamable-HTTP mode (python -m sec_edgar_mcp.server --transport streamable-http --port 9870), explicitly shipped without authentication for private-network use. The only required configuration anywhere is SEC_EDGAR_USER_AGENT, because the SEC requires identifying contact info on automated requests.
Maintained by Stefano Amorelli with a contributor community, an eval suite in CI, and a Zenodo DOI for citation. Trade-offs: AGPL-3.0 copyleft (commercial licensing available separately), and it is unaffiliated with the SEC — it consumes the same public endpoints anyone can, subject to SEC rate expectations.
Tools & capabilities (4)
Company toolsCIK lookup, company info, and company facts.
Filing tools10-K, 10-Q, 8-K retrieval and section extraction.
Financial toolsBalance sheet, income statement, and cash flow, parsed from XBRL.
Insider trading toolsForm 3/4/5 transaction data.
When to use it
- Pull the risk-factors section of a company's latest 10-K into a research summary
- Compare quarterly revenue and cash flow across filings with exact XBRL figures
- Track insider buying/selling from Form 4 transactions for a watchlist
- Verify an AI-generated financial claim against the linked SEC source filing
Quick setup
- 1Set your identification: SEC_EDGAR_USER_AGENT="Your Name (your@email.com)"
- 2Add the Docker config from the README to claude_desktop_config.json (keep the -i flag — it is required for MCP stdio)
- 3Restart your MCP client and ask e.g. 'What was Tesla's operating cash flow last quarter?'
- 4Alternative installs (pip/conda/uv): see sec-edgar-mcp.amorelli.tech/setup/quickstart
- 5For Dify-style platforms: python -m sec_edgar_mcp.server --transport streamable-http --port 9870 (private networks only)
Security notes
Read-only access to public SEC data; no credentials beyond the SEC-required user-agent string with your contact info. The optional streamable-HTTP mode has no authentication built in — the README says to use it only on private networks. AGPL-3.0 license matters if you embed it in a commercial product.
SEC EDGAR MCP FAQ
Is SEC EDGAR MCP free?
Yes. EDGAR data is public and free, and the server is open source (AGPL-3.0) with no API key — you only provide a user-agent string with contact info as the SEC requires. Commercial licensing is offered separately if AGPL doesn't fit your product.
How accurate are the financial numbers?
Financial statements are parsed from the filings' XBRL data, so values are the exact filed figures rather than approximations, and every response includes the SEC filing URL so you can verify. The project also runs a Promptfoo eval suite in CI.
Is this an official SEC tool?
No. The README states it is not affiliated with or endorsed by the U.S. Securities and Exchange Commission; it is a community project by Stefano Amorelli that reads EDGAR's public endpoints.
Alternatives to SEC EDGAR MCP
Compare all alternatives →
Official Stripe server for payments, customers, subscriptions, invoices, and billing via natural language.
Coinbase's AgentKit MCP extension for onchain actions: transfers, swaps, and contract calls.
Alpaca's official trading server for stocks, ETFs, options, and crypto in plain English.