SEC EDGAR
Access public SEC EDGAR company submissions, XBRL facts, filing archives, and filing indexes with the caller identity required by the SEC fair-access policy.
Verdict
Common use cases
- Compare revenue growth across industry peers
- Pull latest 10-Q for quarterly board deck
- Verify insider transaction dates before meeting
- Search XBRL concepts for ESG disclosure trends
- Retrieve full filing text for compliance review
Integration
- Vendor
- SEC EDGAR
- Category
- other
- Auth
- API_KEY
- Tools
- 9
- Composio slug
sec
Tools
- Get SEC Company Concept
Resolve a ticker or CIK and return a bounded page of one company's disclosures for a specific standardized XBRL taxonomy concept.
- Get SEC Company Profile
Resolve a ticker or CIK and return the SEC filer's identity, classification, addresses, tickers, exchanges, and filing-history coverage without returning its filing arrays.
- Get SEC Filing Document
Resolve a filing path and offload one archive file, filing detail page, complete submission, or SGML header while enforcing a caller-selected size cap.
- Get SEC XBRL Frame
Compare one standardized XBRL concept across reporting entities for a calendar period and return a bounded page of matching facts.
- List SEC Company Filings
Resolve a ticker or CIK and return one bounded, normalized page of recent or historical SEC company filings.
- List SEC Filing Documents
Resolve a ticker or CIK and list a bounded page of documents and generated artifacts in one SEC filing archive, including filenames and sizes needed for safe retrieval.
- Search SEC Company Concepts
Resolve a ticker or CIK and search the company's available standardized XBRL concepts. Returns a bounded page of concept metadata and units without returning individual disclosure arrays or values.
- Search SEC Entities
Search SEC companies by ticker, name, or CIK, or mutual funds by ticker, CIK, series ID, or class ID. Returns normalized identifiers for other SEC tools.
- Search SEC Filing Index
Discover and search one daily or quarterly EDGAR master index, returning a bounded page filtered by company, CIK, or form. Quarterly source files over 30 MB are not supported; use a daily index instead.
Setup
Setup guide
- 11. Open your Switchy workspace settings and navigate to the Integrations tab. 2. Find SEC EDGAR in the MCP directory and click Connect. 3. Paste your Switchy API key when prompted—this key meters your requests and does not authenticate with the SEC itself. 4. Click Save and wait for the green confirmation banner. 5. Open any Project and type '@SEC EDGAR' in a message to confirm the connection; you should see the MCP appear in the autocomplete menu. 6. Test the integration by asking '@SEC EDGAR list the last five 10-K filings for Apple' and verify it returns a paginated list with filing dates and accession numbers. 7. If the response is empty or errors, check that your API key is active in workspace settings and that you spelled the ticker correctly.
What teammates see: by default, memories from SEC EDGAR are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Peer Revenue Comparison
@SEC EDGAR pull the 'Revenues' XBRL concept for Tesla, Ford, and GM for Q3 2024 and show me a table comparing their reported figures.
Recent 10-K Summary
@SEC EDGAR get the most recent 10-K filing for Microsoft and summarize the risk factors section and any new business segments mentioned.
Insider Transaction Check
@SEC EDGAR list all Form 4 filings for NVIDIA from the past 30 days and highlight any transactions over one million dollars.
Mutual Fund Lookup
@SEC EDGAR find the Vanguard Total Stock Market Index Fund by ticker VTSAX and pull its most recent N-CSR filing.
XBRL Concept Search
@SEC EDGAR search Apple's available XBRL concepts for anything related to 'debt' and list the concept labels and units.
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@sec-edgar list the last 10 filings for AAPL and highlight any 8-K forms filed in the past 90 days
Apple Inc. (CIK: 0000320193) has filed 10 documents since October 2024. Three 8-K forms appear in this window: a November 1 earnings release (8-K filed 11/01/2024), a December 12 officer departure notice (8-K/A filed 12/12/2024), and a January 3 material agreement disclosure (8-K filed 01/03/2025). The remaining seven filings include quarterly 10-Q reports and proxy materials. All documents are available through the accession numbers provided and can be retrieved individually if you need full text or exhibits.
This example shows the MCP's ability to filter recent filings by form type. The tool returns metadata (form type, filing date, accession number) but does not parse document content—you'll need a follow-up prompt to retrieve and analyze specific exhibits or narrative sections. No authentication required; SEC EDGAR data is public.
@sec-edgar get the full 10-K filing document for Tesla's most recent annual report and cap the download at 5MB
Tesla Inc.'s most recent 10-K (fiscal year ending December 31, 2023, accession number 0001628280-24-002390) is 4.2 MB and falls within your size cap. The complete submission includes the primary 10-K HTML document, 47 exhibits (including audited financials, material contracts, and subsidiary lists), and an XBRL instance document. The filing was accepted by the SEC on January 29, 2024. You can now ask me to summarize specific sections—Item 1A Risk Factors, Item 7 MD&A, or consolidated financial statements—or I can extract exhibit references for deeper review.
This demonstrates safe document retrieval with a user-defined size limit to avoid memory issues. The MCP returns the archive structure but does not automatically parse or summarize content—pair this with a follow-up synthesis prompt to extract insights. Large filings (especially those with embedded images or extensive exhibits) may exceed typical caps, so setting a boundary is critical for production use.
@sec-edgar compare the 'Assets' XBRL concept across all S&P 500 companies for Q4 2023 and show me the top 10 by total assets
For the calendar year 2023 Q4 frame (CY2023Q4), 487 entities reported the us-gaap:Assets concept in their XBRL filings. The top 10 by total assets are: JPMorgan Chase ($3.875T), Bank of America ($3.051T), Citigroup ($2.416T), Wells Fargo ($1.927T), Goldman Sachs ($1.615T), Morgan Stanley ($1.180T), U.S. Bancorp ($675B), PNC Financial ($561B), Truist Financial ($535B), and Capital One ($478B). This frame includes only standardized XBRL facts; companies using non-standard extensions or filing outside the calendar quarter window are excluded from this comparison.
This showcases the MCP's cross-entity XBRL aggregation—a powerful feature for peer benchmarking or sector analysis. The frame is bounded by calendar period and taxonomy concept, so coverage depends on filing schedules and XBRL adoption. Not all filers use the same taxonomy extensions, and some may report assets under custom tags, meaning this list is illustrative rather than exhaustive. Ideal for quantitative screening but should be validated against raw filings for material decisions.
Use-case deep-dives
Why this MCP doesn't help early-stage market-size research
A 4-person startup team building a board deck needs total addressable market estimates for their vertical. The SEC EDGAR MCP won't help: public company filings don't aggregate industry-wide revenue or customer counts, and XBRL frames only compare entities that report the same taxonomy concept (which excludes private companies and non-filers). You'd use this MCP to pull one public competitor's revenue trend as a data point, but for TAM research you need industry reports, census data, or a different MCP that hits market research APIs. If your board deck has a 'competitive landscape' slide citing 10-K risk factors, Get SEC Filing Document is useful. For market sizing, it's not.
Frequently asked
What does the SEC EDGAR MCP do in Switchy?
It pulls public company filings, financial data, and XBRL disclosures directly from the SEC's EDGAR database into your Switchy workspace. You can search companies by ticker or CIK, retrieve specific filings like 10-Ks or 8-Ks, extract standardized financial concepts across reporting periods, and download individual documents from filing archives. It's built for teams analyzing public company disclosures without leaving the AI workspace.
Do I need an API key to use the SEC EDGAR MCP?
Yes. The SEC requires an API key for programmatic access to EDGAR data. You'll need to register for one through the SEC's developer portal and paste it into Switchy's connection settings. The key is free, but the SEC enforces rate limits and requires a valid contact email in case of abuse. One team member with the key can connect it for the whole workspace.
Can it pull real-time stock prices or analyst ratings?
No. This MCP only accesses SEC filings and structured XBRL financial data that companies are legally required to disclose. It won't give you market prices, trading volume, analyst opinions, or anything that isn't in an official SEC filing. For real-time market data, you'd need a separate integration with a financial data provider like Bloomberg or Polygon.
How is this different from reading filings on the SEC website?
The MCP normalizes messy EDGAR data into structured responses you can query with natural language in Switchy. Instead of downloading ZIP archives and parsing XBRL yourself, you ask for specific concepts like revenue or debt across quarters, and the AI returns clean tables. It also enforces size caps on document downloads so you don't accidentally pull gigabytes of exhibits into a chat.
Who on the team should connect the SEC EDGAR MCP?
Anyone doing financial analysis, due diligence, or regulatory research. Typically an analyst, finance lead, or legal team member who already reads 10-Ks and understands XBRL taxonomy. The connection itself is workspace-wide, so once one person adds the API key, everyone in Switchy can query SEC data. It doesn't count against user seats or message limits.