How to monitor SEC EDGAR filings automatically

This workflow checks a company watchlist for new 10-K annual reports, 10-Q quarterly reports, and 8-K material-event filings. It uses the public SEC EDGAR Filings Actor, so it needs no EDGAR API key and produces flat records that can go directly to a spreadsheet, database, Slack workflow, or research agent.

Try the finished search first

Open one of these public tasks to see the output without configuring a new workflow:

The dataset includes the company, form, filing and report dates, accession number, direct SEC filing URL, primary-document URL, and company metadata. Turn on filing details when you also need 8-K item descriptions, exhibit types, or reporting-owner data.

Create the watchlist

In the Actor’s Input tab, use a bounded input like this:

{
  "companies": ["AAPL", "MSFT", "NVDA"],
  "forms": ["10-K", "10-Q", "8-K"],
  "date_from": "2026-09-01",
  "include_amendments": true,
  "include_filing_details": true,
  "max_items": 100
}

Set date_from to the beginning of the period you want the first run to cover. On later runs it is safe to use an overlapping window because accession_number is a stable filing identifier and makes duplicates easy to remove. max_items is a hard output and billing ceiling across the entire company list.

Save the input as an Apify task. In Apify Console, attach that task to a daily schedule and add a run-succeeded webhook if another system should process the resulting dataset immediately.

Keep only genuinely new filings

Store every previously handled accession_number. After each scheduled run:

Read the run’s default dataset.
Discard rows whose accession_number is already in your store.
Route the remaining rows according to form.
Save their accession numbers only after the downstream action succeeds.

For 8-K alerts, include items, item_descriptions, filing_url, and acceptance_datetime in the message. For annual and quarterly research, the useful fields are form, report_date, filing_date, primary_document_url, sic_description, and fiscal_year_end.

Run it from Python

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("captainhandsome/sec-edgar-filings-search").call(run_input={
    "companies": ["AAPL", "MSFT", "NVDA"],
    "forms": ["10-K", "10-Q", "8-K"],
    "date_from": "2026-09-01",
    "include_filing_details": True,
    "max_items": 100,
})
filings = list(client.dataset(run["defaultDatasetId"]).iterate_items())
seen_accessions = set()  # Load this from your persistent store on later runs.
new_filings = [row for row in filings if row["accession_number"] not in seen_accessions]

Replace the example date each time you create a new monitor. Do not put the Apify token in source control.

Ask an AI agent through MCP

With Apify Public Data MCP installed, a one-off check can be as simple as:

Find Tesla’s five latest 8-K filings and return the filing date, accession number, company name, and SEC document URL.

That maps to sec_edgar_filings(ticker="TSLA", form_type="8-K", max_results=5). The scheduled Actor task is the better choice for unattended monitoring; MCP is useful for ad hoc research and summarization.

Cost and limits

The Actor charge is $0.002 per returned filing plus a $0.0005 start charge. A run capped at 100 results therefore has a maximum Actor charge of about $0.2005, and a run returning only three new or overlapping filings is charged for three results. The live Apify Store price is authoritative.

This is a filing monitor, not a real-time market-data feed. SEC publication timing, amendments, and upstream availability determine when a filing appears.

Next steps

This site is open source. Improve this page.