SEC EDGAR Full-Text Search MCP Server

What does SEC EDGAR Full-Text Search MCP Server do?

This Model Context Protocol server gives any MCP-compatible AI client (Claude, Cursor, etc.) direct access to SEC EDGAR Full-Text Search — the SEC's search engine over the full text of filings submitted since 2001. It mirrors every filter available on the search form: keyword query, company/ticker/CIK, filing form type, filed-date range, principal-executive-office or state-of-incorporation location, and SIC industry code. Once a matching filing is found, the server can also list every document inside it and fetch the actual document text, so an agent can go from "find filings mentioning X" all the way to reading the filing itself without leaving the conversation.

Running on the Apify platform means the server is always on (Standby mode), billed per tool call, and benefits from Apify's monitoring, logging, and scheduling.

Why use SEC EDGAR Full-Text Search MCP Server?

• Research and due diligence — search across all companies' 10-Ks, 10-Qs, 8-Ks and other filings for specific language (risk factors, litigation, contract terms).
• Competitive and market intelligence — track what companies in a given industry (SIC code) or state are disclosing.
• Financial and legal analysis pipelines — feed an LLM agent structured filing search results plus the actual document text for summarization or extraction.
• No scraping infrastructure to maintain — the server handles SEC's required User-Agent header, pagination, and result parsing for you.

How to use SEC EDGAR Full-Text Search MCP Server

1. Deploy this Actor to Apify (or link this Git repository) and start a run in Standby mode.

2. Connect your MCP client to https://<your-actor>.apify.actor/mcp using the Streamable HTTP transport, passing your Apify API token as a Bearer token in the Authorization header.

   Add this Actor as an MCP server to Claude Code CLI:

   claude mcp add --transport http sec-edgar https://rl1987--sec-edgar-mcp.apify.actor/mcp \
     --header "Authorization: Bearer $APIFY_API_TOKEN"
   

   Replace $APIFY_API_TOKEN with your Apify API token. Verify it connected with claude mcp list.

3. Call search_edgar_filings with a keyword query and any combination of filters (forms, date range, location, SIC code, entity).

4. Optionally call lookup_edgar_company first to resolve a fuzzy company name/ticker to the exact entity string or CIK.

5. Use get_edgar_filing_index and get_edgar_filing_document to list and read the actual documents behind a search hit.

Input

Configure via the Input tab:

• userAgent (optional) — SEC requires a descriptive User-Agent header identifying your app and contact info (e.g. "MyApp contact@example.com"), per the EDGAR developer FAQ. If left blank, a generic default is used.

Tools

search_edgar_filings

Full-text search with every form filter: query, entity, ciks, forms, date_range (all/10y/5y/1y/custom + start_date/end_date), location_type + location_code, sic, sort (asc/desc/relevance), page, and an optional max_results to auto-paginate beyond the 100-hits-per-page API limit.

lookup_edgar_company

Company/ticker/CIK autocomplete, identical to the typeahead box on the search form.

get_edgar_filing_index

Lists every document contained in a filing (given its CIK and accession number).

get_edgar_filing_document

Fetches a specific filing document by URL and returns its content as Markdown by default (tables — financial statements, beneficial-ownership tables — render as proper Markdown tables, not flattened prose), or raw markup. Supports find/occurrence to jump straight to a section and offset/nextOffset to page through large filings instead of reading from the start.

Output

Each tool call returns structured JSON. Example search_edgar_filings result (truncated):

{
  "totalHits": 698,
  "results": [
    {
      "accessionNumber": "0000320193-23-000106",
      "cik": "0000320193",
      "displayNames": ["Apple Inc.  (AAPL)  (CIK 0000320193)"],
      "form": "10-K",
      "fileDate": "2023-11-03",
      "documentUrl": "https://www.sec.gov/Archives/edgar/data/320193/000032019323000106/a10-kexhibit21109302023.htm",
      "indexUrl": "https://www.sec.gov/Archives/edgar/data/320193/000032019323000106/0000320193-23-000106-index.htm"
    }
  ],
  "facets": { "entities": [], "forms": [], "sics": [], "bizStates": [] }
}

Pricing / Cost estimation

This Actor uses Pay Per Event pricing: a flat fee per tool call (see .actor/pay_per_event.json). There is no charge just for keeping the server in Standby.

Tips

• Wrap multi-word queries in double quotes ("revenue recognition") for exact-phrase matching, matching EDGAR's own search syntax.
• Full-text search only covers filings from 2001 onward; older filings aren't indexed.
• Use max_results sparingly — each page of 100 results is one upstream request; the tool-call fee still applies once per call regardless of how many pages it fetches internally.

FAQ, disclaimers, and support

This Actor scrapes only publicly available data from sec.gov. It is not affiliated with or endorsed by the U.S. Securities and Exchange Commission. Data accuracy depends entirely on SEC EDGAR; always verify against the primary filing before relying on it for legal or financial decisions. Report issues via the Issues tab.