SEC Form 4 Insider Trading: Buying Cluster Signal Scanner

Pricing

Rating

Developer

Actor stats

Categories

Share

Surface scored clusters of insider open-market buying from live SEC Form 4 filings, ranked and ready for research, screening and monitoring.

This Actor watches real-time SEC EDGAR Form 4 insider trading filings, isolates open-market purchases, and groups them by company so you see when several insiders buy the same stock inside a short window. Instead of a raw feed of every grant and option exercise, you get ranked, scored insider buy clusters with a transparent 0-100 score. Insider cluster buying is a widely studied pattern, and this Actor finds it for you automatically.

Why this is different

Most insider trading scrapers hand you raw Form 4 rows: every grant, tax withholding and option exercise mixed together. This Actor delivers the signal instead of the dump.

• Filters to open-market purchases only (transaction code P, acquired), the buys most relevant for screening.
• Detects clusters: at least N different insiders buying the same stock inside your lookback window.
• Scores every cluster 0-100 with a STRONG / MODERATE / WEAK label, weighing the number of distinct insiders, total USD value, and seniority (a CEO or CFO buy counts more than a junior officer).
• Returns clean, ranked JSON ready for a dashboard, model or alert.
• Includes a sec_url with every cluster so you can verify the underlying filings on SEC EDGAR in one click.

It is a self-serve, pay-per-result alternative to subscription platforms like InsiderScore and Fintel.

Who it's for

• Retail quants and algo traders who want a ready-made insider-buying dataset to research and screen, without building an EDGAR parser from scratch.
• Fintech and trading apps that need to drop scored insider signals straight into a product or feed.
• Research desks and newsletter writers who want to surface notable insider clusters on a daily schedule.
• AI agent and automation builders who need a callable tool that returns structured insider trading signals on demand.

Use cases

• Screen for fresh insider buy clusters across the whole market each morning and sort by score to see where multiple insiders are buying at once.
• Feed an AI agent a live "are insiders clustering into ticker X" tool so an LLM can pull current SEC Form 4 signals on demand.
• Monitor a watchlist by running daily and alerting only when a STRONG or MODERATE cluster appears for a company you already track.
• Build a research dataset of historical insider cluster events to backtest or study, with verifiable SEC source links attached to each record.
• Power a newsletter or dashboard that highlights the biggest insider buying clusters of the day, ranked by dollar value and seniority.

Input

Output

Each dataset item is one scored insider-buy cluster. Example using real values detected by this Actor in live SEC data:

{
"ticker":"AADX",
"issuer":"Applied Aerospace & Defense, Inc.",
"cik":"0002118195",
"num_insiders":4,
"total_value":1170000.0,
"num_buys":4,
"avg_seniority":0.6,
"score":71.7,
"label":"MODERATE",
"buys":[
{"owner":"McRae Jeffrey L.","role":"Chief Financial Officer",
"shares":25000,"price":20.0,"value":500000.0,"date":"2026-06-04"},
{"owner":"Katzman James C","role":"Director",
"shares":25000,"price":20.0,"value":500000.0,"date":"2026-06-04"}
],
"sec_url":"https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=2118195&type=4&owner=include&count=40"
}

In this real cluster, 4 insiders including the CFO bought $1.17M of the same stock on the same day, scoring 71.7 (MODERATE). The sec_url links to that company's Form 4 filings on SEC EDGAR so you can confirm every row.

How the score works

The score is transparent and interpretable - no black box:

score = insiders (up to 55 pts) -> min(num_insiders, 6) / 6 * 55
 + dollar size (up to 30 pts) -> min(log10(total_value) / 7, 1) * 30(caps around $10M)
 + seniority (up to 15 pts) -> avg_seniority * 15
labels: STRONG >=75 MODERATE >=50 WEAK <50

More distinct insiders, more dollars, and more senior buyers (CEO 1.0, CFO 0.9, President 0.8, COO / Chairman 0.7, Director 0.5, other officer 0.3) all push the score up. The dollar component is logarithmic, so a $1M buy counts but does not swamp the score versus a $1k buy.

Use with AI agents and automation

This runs as a standard Apify Actor, so it is callable as a live tool by AI agents and LLMs, not just a one-off script.

• AI agent frameworks: wire it into LangChain or LlamaIndex as a tool an agent calls to fetch current insider trading signals while it reasons.
• Apify MCP server: expose this Actor over the Model Context Protocol so assistants like Claude, ChatGPT and Cursor can call it directly and get structured insider cluster signals back in the conversation.
• No-code automation: trigger it from Zapier or Make to push fresh clusters into a sheet, CRM, Slack channel or email.
• Webhooks and pipelines: fire an Apify webhook on each run to hand new clusters to your own backend.
• JSON on demand or on schedule: run it ad hoc through the API, or set a daily schedule so signals accumulate automatically.

Pricing

Pay-per-event: you pay only for each delivered cluster signal. There is no subscription, and no charge when a run finds no clusters. Run it on a daily schedule and you only pay for real, scored signals delivered to your dataset.

Data source and compliance

• Source is SEC EDGAR Form 4 filings: public-domain U.S. regulatory data that is free and legal to use.
• No personal or contact data is collected - only public corporate filing facts (issuer, the insider name and role as disclosed in the filing, and the transaction details).
• The Actor respects SEC fair-access rules with a descriptive User-Agent and rate-limited requests well under the 10 requests-per-second cap.

FAQ

Can an AI agent call this automatically? Yes. Because it is a standard Apify Actor, you can expose it through the Apify MCP server and let agents in Claude, ChatGPT or Cursor call it directly, or wire it into LangChain or LlamaIndex as a callable tool.

Why are some runs empty? Open-market insider buy clusters are genuinely rare, which is exactly why they work as a signal. Run daily on a schedule and let signals accumulate, or widen daysLookback.

Does it include sells? No. It covers cluster buying only (transaction code P). Sells, grants and option exercises are intentionally excluded.

Is it real-time? Form 4 filings appear on EDGAR within seconds of submission, and this Actor reads the latest daily filings on each run.

How do I tune the strictness? Raise minInsiders for tighter clusters, set minClusterValue to ignore small-dollar clusters, and increase daysLookback to widen the window. maxFilings caps how many filings are scanned per run.

Can I verify the underlying filings? Yes. Every cluster includes a sec_url that links to that company's Form 4 filings on SEC EDGAR so you can check the raw source.

Keywords: SEC Form 4, insider trading, insider buying, insider cluster, cluster buying signal, SEC EDGAR scraper, insider buy alerts, insider buying data, insider transactions API, insider screening data, quant research data.

Note: Provided as data for research, screening and monitoring, not investment advice. Historical patterns do not guarantee future results.