SEC EDGAR MCP server
A maintained MCP server that connects agents to SEC EDGAR filings, XBRL financials, and insider trading with exact numeric precision.
SEC EDGAR MCP is an actively maintained community server that gives an AI agent direct, citable access to the U.S. Securities and Exchange Commission's EDGAR database of public-company filings. Built on the edgartools library, it resolves a ticker to a CIK, pulls company info and the full set of company facts, lists recent filings filtered by form type and date, and retrieves filing content and specific sections — including a dedicated 8-K analyzer for material events. Every response carries the underlying SEC filing URL so claims can be verified against the primary source.
For numbers it parses XBRL directly, returning income statements, balance sheets, and cash-flow data with exact figures rather than rounded estimates, plus segment data, key metrics, period-over-period comparisons, and tools to discover the available XBRL concepts for a company. A full insider-trading suite reads Form 3/4/5 transactions, summarizes activity, extracts Form 4 detail, and analyzes insider sentiment over time. The server is published as the sec-edgar-mcp Python package, runs locally over stdio (with an optional streamable-HTTP transport), and requires a SEC_EDGAR_USER_AGENT identifying you to the SEC. It is AGPL-3.0 licensed with commercial licensing available.
Quick install
Copy-paste configs are provided for all 8 supported clients. Pick your client below.
Available tools
| Tool | Description |
|---|---|
| get_cik_by_ticker | Resolves a stock ticker to its SEC Central Index Key (CIK). |
| get_company_info | Returns company information for a given identifier. |
| search_companies | Searches for companies matching a query. |
| get_company_facts | Returns the full set of standardized company facts from SEC data. |
| get_recent_filings | Lists a company's recent filings, filterable by form type and date range. |
| get_filing_content | Retrieves the content of a specific filing by accession number, with offset and length controls. |
| analyze_8k | Analyzes an 8-K current report to surface the material events it discloses. |
| get_filing_sections | Extracts specific sections from a filing by form type. |
| get_financials | Returns income statement, balance sheet, and cash-flow data parsed from XBRL. |
| get_segment_data | Returns business or geographic segment data for a company. |
| get_key_metrics | Returns selected key financial metrics for a company. |
| compare_periods | Compares a metric across a start and end year to show growth. |
| discover_company_metrics | Discovers the available XBRL concepts and metrics for a company. |
| get_xbrl_concepts | Returns the values of specific XBRL concepts from a filing. |
| discover_xbrl_concepts | Lists the XBRL concepts available in a filing, optionally filtered by namespace. |
| get_insider_transactions | Returns insider trading transactions (Form 3/4/5) for a company. |
| get_insider_summary | Summarizes insider trading activity over a time window. |
| get_form4_details | Returns detailed transaction data from a specific Form 4 filing. |
| analyze_form4_transactions | Analyzes recent Form 4 transactions for a company. |
| analyze_insider_sentiment | Analyzes insider buying versus selling sentiment over a period. |
| get_recommended_tools | Recommends which tools to use for analyzing a given SEC form type. |
Required configuration
- SEC_EDGAR_USER_AGENTRequired
Your name and email identifying you to the SEC, required by EDGAR's fair-access policy.
What you can do with it
Read a company's financials from the source
The agent resolves a ticker, pulls XBRL-parsed income statement, balance sheet, and cash flow with exact figures, and includes the SEC filing URL so every number can be verified.
Track insider activity
Pull Form 3/4/5 transactions, summarize buying versus selling, and analyze insider sentiment over months to flag notable executive moves.
FAQ
- Is it free?
- Yes. The server is open source under the AGPL-3.0 license and SEC EDGAR data is free and public. You only need to set a SEC_EDGAR_USER_AGENT with your name and email. Commercial licensing is available from the author for closed-source use.
- Does it support remote/OAuth?
- It runs locally over stdio by default and can also run over a streamable-HTTP transport for platforms that need it. EDGAR data is public, so it authenticates only with the required user-agent header rather than OAuth.
- How accurate are the numbers?
- Financial data is parsed directly from each filing's XBRL with exact figures, not rounded estimates, and every response includes the source SEC filing URL so an agent or human can verify against the original document.