SEC Financial Intelligence MCP Server

In one sentence

SEC Financial Intelligence MCP Server is a Model Context Protocol (MCP) server on Apify that lets AI agents query SEC EDGAR filings, insider trading, institutional 13F holdings, and congressional stock trades, and returns one unified investment-intelligence report with cross-domain BULLISH/BEARISH/NOTABLE signals.

Category: SEC filing search tool. Investment research API. AI agent data source. Primary use case: Run one query and get insider, institutional, and congressional activity for a company with signal detection. Can also be used for single-source lookups like full-text EDGAR search or XBRL financial extraction.

Also known as: SEC EDGAR MCP, insider trading MCP, 13F holdings API, congressional stock trades tool

What this actor does

• What it is: A standby MCP server exposing 7 SEC financial intelligence tools to AI agents over the /mcp endpoint.
• What it checks: SEC EDGAR full-text filings, Form 4 insider transactions, Form 13F-HR institutional holdings, and STOCK Act congressional trades.
• What it returns: Structured JSON. Its headline tool, sec_investment_intelligence, returns a single profile with filing summaries, insider buy/sell direction, congressional activity, and a list of detected signals.
• What it does NOT do: It does not give investment advice, does not predict prices, and does not cover short positions, options, or bond holdings in 13F data.
• Who it's for: AI agent builders, equity researchers, hedge fund analysts, financial journalists, and developers building investment tools.

SEC Financial Intelligence MCP Server is an Apify actor that connects AI agents to four SEC data domains through one MCP endpoint. Instead of wiring separate integrations for EDGAR, insider trading, 13F filings, and congressional disclosures, an agent calls a single tool and receives a finished, cross-referenced view. It functions as a SEC investment research API, returning structured records ready for agent reasoning.

The differentiator is the composite tool. sec_investment_intelligence queries EDGAR filings, insider activity, and congressional trades in parallel, then emits plain signal lines such as "BULLISH: net insider buying" or "NOTABLE: congressional purchases in the last 90 days." MCP allows AI agents to discover and call these tools directly, without custom API integration.

To research a company across SEC data, an agent calls sec_investment_intelligence with a ticker. It returns recent filings, XBRL financials, insider buy/sell totals, congressional trades, and a signal list in one response. For cheaper single-domain lookups, five focused tools cover search, company analysis, insiders, institutions, and Congress separately.

In short: one MCP call returns insider, institutional, and congressional SEC activity for any public company, with signals already detected.

What it is: an MCP server for SEC financial intelligence. Who it's for: AI agents, equity analysts, and developers building investment research workflows. When to use it: when an agent needs a finished cross-domain SEC profile, not raw filing-by-filing records.

What it does — exposes 7 SEC tools (search, analyze, insider, institutional, congressional, composite, list) to MCP clients. Best for — agent-driven investment due diligence and SEC data retrieval. Pricing — pay-per-event, $0.05 to $0.15 per tool call; sec_list_sources is free. Output — structured JSON over the MCP protocol.

Best when the user wants an answer, not raw filings — SEC Financial Intelligence MCP Server returns finished cross-domain profiles rather than source-by-source records.

Key limitation: 13F institutional data covers long equity only. Congressional trade amounts are reported in ranges, not exact values. What it is not: Not an investment advisor, not a price-prediction model, and not a replacement for primary SEC filings. Does not include: Short positions, options, bonds, real-time intraday prices, or non-US disclosures. Results may be incomplete when: a company has no matching EDGAR record, an investor has filed no recent 13F, or no STOCK Act disclosures fall in the requested window.

What you get from one call

Input: sec_investment_intelligence with { "company": "NVDA", "days_back": 90 } Returns:

• Company metadata (CIK, SIC, tickers) and recent filings grouped by type
• XBRL financials (revenue, net income, assets) when available
• Insider trading summary: purchases vs sales, total buy and sell value, net direction
• Congressional trades touching the ticker in the window
• A signal list (BULLISH / BEARISH / NOTABLE / ACTIVE) derived from the combined data

Typical time to first result: seconds to a couple of minutes for the composite tool (it runs up to 4 sub-actors in parallel). Typical time to integrate: minutes — point any MCP client at the /mcp URL.

What makes this different

• One call across four SEC domains — insider, institutional, and congressional activity plus filings come back in a single composite tool with signals attached.
• Signal detection built in — sec_investment_intelligence flags net insider buying, heavy selling, congressional activity, and elevated 8-K volume so an agent reads conclusions, not raw rows.
• Honest scope — 13F long-equity-only and congressional range-reporting limits are stated up front, not buried.

If you were building this yourself, you would need to integrate the SEC EDGAR full-text search API, parse XBRL financial statements, normalize Form 4 transaction codes, diff 13F filings quarter over quarter, and scrape House and Senate STOCK Act disclosures.

Tools

The composite tool is the recommended first call when a task involves a public company. The five focused tools are faster and cheaper for single-domain lookups.

Tool	Price	Description
sec_investment_intelligence	$0.15	Composite report. Runs EDGAR, insider, and congressional sources in parallel and returns a unified profile with cross-domain BULLISH/BEARISH/NOTABLE signal analysis. Up to 4 actor calls.
sec_search_filings	$0.05	Full-text search across SEC EDGAR filings (10-K, 10-Q, 8-K, S-1, DEF 14A, Form 4, SC 13D/G, 20-F, 6-K) by keyword, company, or ticker. Returns metadata with direct document links.
sec_analyze_company	$0.10	Company filing analysis with structured XBRL financial extraction (revenue, net income, assets), CIK / SIC / tickers, and recent filings.
sec_insider_trading	$0.05	Form 4 insider activity (purchases, sales, grants, exercises) by officers, directors, and 10% owners. Aggregates buy vs sell counts and value.
sec_institutional_holdings	$0.05	Form 13F-HR institutional holdings (hedge funds, mutual funds, managers) with quarter-over-quarter position changes. Long equity only.
sec_congressional_trades	$0.05	STOCK Act stock-trade disclosures by US House and Senate members. Amounts reported in ranges.
sec_list_sources	Free	Lists available data sources and capabilities. No charge.

Data sources

Source	Powers	Coverage
SEC EDGAR EFTS full-text search	sec_search_filings	All SEC filing types by keyword, company, or ticker
SEC EDGAR + XBRL company financials	sec_analyze_company	Filing metadata plus structured revenue, income, and asset data
SEC Form 4 (insider)	sec_insider_trading	Purchases, sales, grants, and option exercises by insiders
SEC Form 13F-HR (institutional)	sec_institutional_holdings	Quarterly holdings from 5,000+ institutional filers (long equity only)
House / Senate STOCK Act disclosures	sec_congressional_trades	Stock trades by members of Congress, amounts in ranges

Quick answers

What is it? An MCP server that gives AI agents SEC financial intelligence across filings, insider trades, institutional holdings, and congressional disclosures. How do I use it across SEC sources at once? Call sec_investment_intelligence with a ticker or company name to query EDGAR, insider, and congressional data in parallel and get back signals. What data sources does it use? SEC EDGAR full-text and XBRL, SEC Form 4, SEC Form 13F-HR, and House/Senate STOCK Act disclosures. What does it return? Structured JSON over MCP: filing summaries, XBRL financials, insider buy/sell direction, congressional activity, and detected signals. How much does it cost? Pay-per-event, $0.05 to $0.15 per tool call. sec_list_sources is free. What are the main limits? 13F covers long equity only. Congressional amounts are reported as ranges. Results can be empty when no matching filings exist.

At a glance

Quick facts:

• Input: A company name, ticker, CIK, investor name, or Congress member name, depending on the tool.
• Output: Structured JSON over the MCP protocol.
• Pricing: Pay-per-event, $0.05 to $0.15 per tool call; sec_list_sources free.
• Tools: 7 (1 composite, 5 focused, 1 free source-listing).
• Sources: 5 SEC and government data domains.
• Transport: Standby MCP server, streamable HTTP at /mcp.

Input → Output:

• Input: company / ticker / investor / member name
• Process: the server calls the matching SEC sub-actor(s), aggregates results, and detects signals for the composite tool
• Output: structured JSON ready for agent reasoning

Problems this solves:

• How to give an AI agent SEC filing access without building an EDGAR integration
• How to see insider buying vs selling for a company in one call
• How to track which institutions are increasing or exiting a position
• How to check whether members of Congress traded a given ticker

Best fit: AI agent investment research, due diligence workflows, equity screening, financial journalism. Not ideal for: Real-time price feeds, options/short data, non-US disclosures, automated trading execution. Does not include: Short positions, options, bonds, intraday prices, or investment advice.

Common questions this actor answers:

• Did insiders buy or sell recently? sec_insider_trading aggregates Form 4 purchases vs sales and value.
• What is a hedge fund holding now vs last quarter? sec_institutional_holdings returns 13F positions with quarter-over-quarter changes.
• Did Congress trade this stock? sec_congressional_trades filters STOCK Act disclosures by ticker and chamber.
• What is the full picture for this company? sec_investment_intelligence combines all of the above with signals.

What is a SEC financial intelligence MCP?

A SEC financial intelligence MCP is a Model Context Protocol server that turns SEC and government disclosure data into tools an AI agent can call directly. Basic SEC tools return one feed; SEC Financial Intelligence MCP Server combines EDGAR filings, Form 4 insider trades, Form 13F-HR holdings, and STOCK Act congressional disclosures behind one endpoint, and its composite tool reads conclusions from the combined data rather than handing back raw rows.

Features

SEC Financial Intelligence MCP Server exposes 7 registered MCP tools backed by 5 SEC and government data sources. The composite tool detects buying and selling signals across insiders, institutions, and Congress in a single response, while focused tools cover each domain independently.

Search and analysis

• Full-text EDGAR search — query 10-K, 10-Q, 8-K, S-1, DEF 14A, Form 4, SC 13D/G, 20-F, and 6-K filings by keyword, company, or ticker, with direct document links.
• XBRL financial extraction — pull structured revenue, net income, and assets plus CIK, SIC, and ticker metadata for a named company.

Signal and activity tracking

• Insider direction — aggregate Form 4 purchases, sales, grants, and exercises into buy vs sell counts and dollar value.
• Institutional position changes — surface 13F-HR holdings with new positions, exits, increases, and decreases quarter over quarter.
• Congressional trades — filter STOCK Act disclosures by member, ticker, chamber, and transaction type.

Composite intelligence

• Cross-domain signals — sec_investment_intelligence runs sources in parallel and emits BULLISH, BEARISH, NOTABLE, and ACTIVE signal lines with a net insider direction summary.
• Free source discovery — sec_list_sources returns the source map and capabilities at no charge, so an agent can plan calls before spending.

Use cases for SEC financial intelligence

Best for AI agent due diligence

Use when an agent needs a full SEC picture of a company before answering. One sec_investment_intelligence call returns filings, financials, insider direction, and congressional activity. Used by analyst-assistant agents to produce company briefs without chaining multiple integrations.

Best for insider signal monitoring

Use when tracking executive buying and selling. sec_insider_trading aggregates Form 4 activity into net direction and value, useful for spotting cluster insider buying within a short window.

Best for institutional position tracking

Use when following what hedge funds and asset managers hold. sec_institutional_holdings returns 13F positions with quarter-over-quarter changes for a named investor. Key outputs: holdings, position values, new/exited positions.

Best for political trading research

Use when investigating whether members of Congress traded a regulated company. sec_congressional_trades filters STOCK Act disclosures by member, ticker, and chamber. Key outputs: member, chamber, transaction type, disclosure date, filing link.

How to connect SEC Financial Intelligence MCP Server

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "sec-financial-intelligence": {
      "url": "https://sec-financial-intelligence-mcp.apify.actor/mcp"
    }
  }
}

Programmatic (HTTP)

List the available tools:

curl -X POST https://sec-financial-intelligence-mcp.apify.actor/mcp \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_APIFY_TOKEN" \
  -d '{"jsonrpc":"2.0","method":"tools/list","id":1}'

Call the composite tool:

curl -X POST https://sec-financial-intelligence-mcp.apify.actor/mcp \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_APIFY_TOKEN" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "id": 2,
    "params": {
      "name": "sec_investment_intelligence",
      "arguments": { "company": "NVDA", "days_back": 90 }
    }
  }'

First run tips

• Start with sec_list_sources — it is free and returns the source map, so an agent can plan which paid tools to call.
• Use focused tools for single questions — sec_insider_trading or sec_congressional_trades are cheaper and faster than the composite tool when you only need one domain.
• Reserve the composite tool for full profiles — sec_investment_intelligence runs up to 4 sub-actors and costs $0.15, so use it when you genuinely want everything at once.
• Expect range values for Congress — STOCK Act amounts come back as ranges like "$1,001-$15,000"; exact figures live in the linked PDF disclosures.
• Empty is not an error — a company with no matching filings or an investor with no recent 13F returns an explicit empty/not-found result.

Tool input parameters

Each tool accepts a small set of arguments. Key parameters:

Tool	Key parameters
sec_search_filings	query, form_type, date_from, date_to, max_results (1-200)
sec_analyze_company	company, filing_types, include_financials, max_filings (1-500), date range
sec_insider_trading	company, transaction_type, date range, min_value, max_results
sec_institutional_holdings	investor, include_changes, max_holdings (1-10000), min_position_value
sec_congressional_trades	member, ticker, chamber, transaction_type, days_back (1-730), max_results
sec_investment_intelligence	company, days_back (1-365), include_financials

Output example

A sec_investment_intelligence response is structured JSON:

{
  "company": { "name": "NVIDIA Corp", "cik": "0001045810", "sic": "3674", "tickers": ["NVDA"] },
  "analysisperiod": { "from": "2026-02-20", "to": "present", "daysBack": 90 },
  "signals": [
    "BULLISH: 4 insider purchases (4 buys vs 1 sells) — net buying signal",
    "NOTABLE: 2 congressional purchase(s) in last 90 days",
    "ACTIVE: 6 8-K current event reports filed — high corporate activity"
  ],
  "filingsSummary": {
    "totalFilings": 18,
    "byType": { "10-Q": 1, "8-K": 6, "4": 11 },
    "mostRecent": [
      { "type": "8-K", "date": "2026-05-08", "description": "Results of Operations" }
    ]
  },
  "financials": { "revenue": 130497000000, "netIncome": 72880000000, "totalAssets": 111601000000 },
  "insiderTrading": {
    "totalTrades": 5, "purchases": 4, "sales": 1,
    "totalBuyValue": 2870000, "totalSellValue": 410000,
    "netDirection": "NET BUYING",
    "topInsiders": [
      { "name": "Jensen Huang", "title": "President & CEO", "type": "Purchase", "value": 1200000, "date": "2026-04-30" }
    ]
  },
  "congressionalActivity": {
    "totalTrades": 2,
    "trades": [
      { "member": "Jane Doe", "chamber": "house", "type": "Purchase", "disclosure": "2026-04-15", "asset": "NVIDIA Corp", "filingUrl": "https://disclosures-clerk.house.gov/..." }
    ]
  }
}

Output fields

Field	Type	Description
company	object	CIK, SIC, name, and tickers from EDGAR
signals	array	Detected BULLISH / BEARISH / NOTABLE / ACTIVE lines
filingsSummary.byType	object	Filing counts grouped by form type
financials	object/null	XBRL revenue, net income, assets (null when unavailable)
insiderTrading.netDirection	string	NET BUYING, NET SELLING, or NEUTRAL
insiderTrading.topInsiders	array	Insider name, title, transaction type, value, date
congressionalActivity.trades	array	Member, chamber, type, disclosure date, asset, filing URL

Focused tools return their domain payload directly: sec_search_filings returns { total, filings }, sec_insider_trading returns aggregated buy/sell counts and a trades array, and so on.

How much does it cost to query SEC data?

SEC Financial Intelligence MCP Server uses pay-per-event pricing — you pay only when a tool is called, and platform compute is included. sec_list_sources is free.

Tool call	Price
sec_list_sources	Free
sec_search_filings	$0.05
sec_insider_trading	$0.05
sec_institutional_holdings	$0.05
sec_congressional_trades	$0.05
sec_analyze_company	$0.10
sec_investment_intelligence	$0.15

Example monthly cost:

Scenario	Calls	Approx. total
Quick test	1 composite call	$0.15
Light research	20 focused calls	~$1.00
Daily monitoring	100 mixed calls	~$8.00
Heavy due diligence	50 composite calls	$7.50

Set a spending limit in your Apify account to cap usage. The Apify Free plan includes $5 of monthly credits, enough for multiple queries across any tools.

How SEC Financial Intelligence MCP Server works

Mental model: Agent calls a tool → server calls the matching SEC sub-actor(s) → results are aggregated → signals are detected (composite tool) → structured JSON returns over MCP.

Standby MCP transport

The actor runs in standby mode and serves the MCP protocol over streamable HTTP at /mcp. An MCP client connects to the URL, lists tools, and calls them. The server cold-starts on the first request and shuts down after a period of inactivity to release platform compute.

Tool routing

Each tool maps to a dedicated SEC sub-actor. Focused tools call one sub-actor; sec_investment_intelligence calls EDGAR analysis, insider trading, and congressional trades in parallel, then merges the results.

Signal detection

The composite tool applies threshold rules to the merged data: net insider buying with three or more purchases flags BULLISH, heavy selling flags BEARISH, congressional trades flag NOTABLE, and five or more recent 8-K reports flag ACTIVE. When nothing crosses a threshold, it returns a no-significant-signals line.

Limitations

• 13F is long equity only — institutional holdings exclude short positions, options, and bonds.
• Congressional amounts are ranges — STOCK Act disclosures report value bands, not exact figures; precise data is in the linked PDFs.
• 13F filing lag — 13F-HR filings arrive about 45 days after quarter end, so holdings are not real-time.
• US scope — coverage is SEC and US congressional disclosures; non-US filings are out of scope.
• No price data — the server does not return intraday or historical market prices.
• Empty results — a company, investor, or member with no matching filings returns an explicit empty or not-found response, not an error.
• Not advice — outputs are informational and do not constitute investment recommendations.

Combine with other Apify actors

Actor	How to combine
Company Deep Research	Pair SEC signals with broader company intelligence reports
Trustpilot Review Analyzer	Add consumer-reputation signals to a financial profile
Website Tech Stack Detector	Profile a company's technology alongside its filings
WHOIS Domain Lookup	Verify corporate domains during due diligence
Website Content to Markdown	Feed filing-linked pages into an LLM or RAG pipeline

Responsible use

• SEC Financial Intelligence MCP Server accesses publicly available SEC filings and government disclosure data. It does not bypass authentication, CAPTCHAs, or access restricted content.
• Users are responsible for ensuring their use complies with applicable laws and platform terms, including securities regulations in their jurisdiction.
• Outputs are informational only and are not investment advice. Do not use the data for market manipulation or other unauthorized purposes.
• For guidance on web scraping legality, see Apify's guide.

FAQ

What is the difference between a basic SEC API and this MCP server? A basic API returns one data feed. SEC Financial Intelligence MCP Server combines EDGAR filings, insider trades, 13F holdings, and congressional disclosures, and its composite tool returns detected signals instead of raw rows.

Can I query all SEC sources in one call? Yes. sec_investment_intelligence runs EDGAR, insider, and congressional sources in parallel and returns a unified profile with signals.

Does this server give investment advice? No. It returns factual SEC and disclosure data with signal flags. It does not recommend trades or predict prices.

How does the signal detection work? The composite tool applies threshold rules to merged data, flagging net insider buying as BULLISH, heavy selling as BEARISH, congressional activity as NOTABLE, and high 8-K volume as ACTIVE.

Why are congressional trade amounts shown as ranges? STOCK Act disclosures report value bands like "$1,001-$15,000" rather than exact amounts. Exact figures are in the linked filings.

Does 13F data include short positions or options? No. Form 13F-HR covers long equity positions only.

How current is the institutional holdings data? 13F-HR filings are submitted quarterly, roughly 45 days after quarter end, so holdings reflect that filing cadence rather than live positions.

How do I connect an AI agent to it? Point any MCP client, such as Claude Desktop, at https://sec-financial-intelligence-mcp.apify.actor/mcp and list the tools.

What does it cost to try? A single sec_investment_intelligence call is $0.15; focused tools are $0.05 to $0.10; sec_list_sources is free. The Apify Free plan covers multiple queries.

Can I filter insider trades by type? Yes. sec_insider_trading accepts a transaction type filter for purchases, sales, grants, or exercises.

Is it a practical alternative to wiring SEC integrations by hand? Yes, for agent workflows it replaces separate EDGAR, Form 4, 13F, and STOCK Act integrations with one MCP endpoint.

Is it legal to use this data? It accesses only publicly available SEC and US congressional disclosures. Legality depends on your jurisdiction and intended use; consult legal counsel for your specific case. See Apify's guide.

Help us improve

If you encounter issues, you can help us debug faster by enabling run sharing in your Apify account:

1. Go to Account Settings > Privacy
2. Enable Share runs with public Actor creators

This lets us see your run details when something goes wrong, so we can fix issues faster. Your data is only visible to the actor developer, not publicly.

Support

Found a bug or have a feature request? Open an issue in the Issues tab on this actor's page. For custom solutions or enterprise integrations, reach out through the Apify platform.