Insider Trades API — Python SDK

Official Python SDK for the Insider Trades API — real-time SEC Form 3, 4, and 5 insider transaction data.

Using Claude, Cursor, ChatGPT, or another AI assistant? Use the insider-trades-mcp package instead — it connects your AI tool directly to the Insider Trades API with no code required.

Using JavaScript or TypeScript? See the insider-trades-api Node.js SDK on npm.

Installation

pip install insider-trades-api

Requires Python 3.8+. No runtime dependencies.

Quick Start

import os
from insider_trades_api import InsiderTradesAPI

client = InsiderTradesAPI(api_key=os.environ["INSIDER_TRADES_API_KEY"])

# Get recent insider purchases for Apple
result = client.get_insider_transactions(
    issuer_ticker="AAPL",
    transaction_types="purchase",
    limit=10,
)

for tx in result["insiderTransactions"]:
    owner = tx.get("reportingOwnerInformation", {})
    print(owner.get("name"), tx.get("purchaseAmount"))

If you omit api_key, the client reads INSIDER_TRADES_API_KEY from the environment.

Get your API key at insidertrades.us.

---

API Reference

For the full reference including all parameters, response schemas, and plan limits, see insidertrades.us/docs.

InsiderTradesAPI(api_key=None, *, timeout=30.0)

Argument	Type	Required	Description
api_key	str	Yes	Your API key. Falls back to INSIDER_TRADES_API_KEY env var
timeout	float	No	Per-request socket timeout in seconds (default 30.0)

---

get_insider_transactions(**params)

Retrieve SEC insider transactions from Form 3, 4, and 5 filings. All arguments are keyword-only. Returns the decoded JSON response as a dict with insiderTransactions, count, recordedTimeInUtc, hasMore, and (when hasMore is True) nextCursor.

result = client.get_insider_transactions(
    issuer_ticker="AAPL",
    transaction_types=["purchase", "sale"],
    filing_date_in_est_start_date="2026-01-01",
    filing_date_in_est_end_date="2026-01-31",
    limit=50,
)
print(result["count"], len(result["insiderTransactions"]))

Parameter	Type	Description
filing_id	str	Direct lookup by filing ID
issuer_ticker	str	Filter by ticker symbol, e.g. "AAPL"
issuer_cik	str	Filter by issuer CIK
reporting_owner_cik	str	Filter by insider CIK
accession_number	str	Filter by SEC accession number
form_types	str | Sequence[str]	Form types: "3", "4", "5" (default: all)
transaction_types	str | Sequence[str]	See transaction types below
owner_roles	str | Sequence[str]	"director", "officer", "tenPercentOwner"
officer_titles	str | Sequence[str]	Normalized titles: "CEO", "CFO", "COO", etc.
filing_date_in_est_start_date	str	Filing date start YYYY-MM-DD (Eastern Time)
filing_date_in_est_end_date	str	Filing date end YYYY-MM-DD (Eastern Time)
period_of_report_start_date	str	Period of report start YYYY-MM-DD
period_of_report_end_date	str	Period of report end YYYY-MM-DD
min_total_amount	float	Minimum total dollar value
max_total_amount	float	Maximum total dollar value
fieldset	"minimal" | "standard" | "full"	Control response size (see below)
limit	int	Page size, default 100, max 750
cursor	str	Opaque pagination cursor — pass back the nextCursor from a previous response (see Pagination)

Transaction types

Value	SEC Code	Description
"purchase"	P	Open market purchase
"sale"	S	Open market sale
"grant"	A	Grant or award
"gift"	G	Bona fide gift
"exercise"	M	Option exercise (Rule 16b-3)
"derivativeExercise"	O, X	Derivative exercise
"disposition"	D	Disposition to issuer
"discretionary"	I	Discretionary transaction
"derivativeConversion"	C	Conversion of derivative
"derivativeExpiration"	E, H	Expiration of derivative
"smallAcquisition"	L	Small acquisition (Rule 16a-6)
"inheritance"	W	Inheritance
"equitySwap"	K	Equity swap
"tender"	U	Tender offer / change of control
"other"	J	Other acquisition / disposition

Fieldsets

Fieldset	Fields included
"minimal"	ID, issuer, owner, key amounts, dates
"standard"	All of minimal + all aggregates, boolean flags, links, AI summary
"full"	All of standard + raw transaction arrays (nonDerivativeTransactions, derivativeTransactions, holdings)

Default (no fieldset): all stored fields are returned. minimal and standard produce significantly smaller payloads — recommended for bulk queries.

Sample response:

{
  "insiderTransactions": [
    {
      "id": "97e823b724619e921c68ba73b2f95f3b015fb83cc05666320b6ddf34b97cdcbe",
      "formType": "4",
      "accessionNumber": "0001140361-26-023363",
      "issuerCik": "320193",
      "issuerInformation": {
        "cik": "0000320193",
        "tradingSymbol": "AAPL",
        "companyName": "Apple Inc."
      },
      "reportingOwnerCik": "1214128",
      "reportingOwnerInformation": {
        "cik": "0001214128",
        "name": "LEVINSON ARTHUR D",
        "city": "CUPERTINO",
        "state": "CA",
        "isDirector": true,
        "isOfficer": false
      },
      "ownerRoles": "director",
      "filingDateInEst": "2026-05-29",
      "periodOfReport": "2026-05-27",
      "totalAmount": 15551000.0,
      "saleAmount": 15551000.0,
      "saleCount": 1,
      "netSharesAcquired": -50000.0,
      "netTradingAmount": -15551000.0,
      "hasSales": "true",
      "hasLateTransactions": false,
      "aff10b5One": false,
      "summary": "Arthur D. Levinson sold 50,000 shares of Apple Inc. common stock at $311.02 per share on May 27, 2026.",
      "linkToFilingDetail": "https://www.sec.gov/Archives/edgar/data/1214128/000114036126023363/0001140361-26-023363-index.htm"
    }
  ],
  "count": 1,
  "recordedTimeInUtc": "2026-05-29T22:30:00Z",
  "hasMore": false
}

---

iter_insider_transactions(**params)

Auto-paginating generator that yields individual transactions across all pages, following nextCursor for you. Accepts the same filters as get_insider_transactions (the cursor is managed internally). Iteration is lazy — pages are fetched on demand.

# Walk an entire year without managing cursors yourself.
for tx in client.iter_insider_transactions(
    filing_date_in_est_start_date="2025-01-01",
    filing_date_in_est_end_date="2025-12-31",
    limit=750,  # larger page size = fewer requests against your quota
):
    print(tx["id"], tx.get("totalAmount"))

---

Pagination

A single response holds at most limit records (capped at 750 to stay within the API's response-size limit). When more results remain, the response sets "hasMore": true and includes an opaque nextCursor.

Easiest: let the SDK walk pages for you with iter_insider_transactions.

Manual: pass nextCursor back as cursor, keeping the same filter arguments. A cursor minted for one set of filters is rejected (HTTP 400) if replayed against different ones.

cursor = None
while True:
    page = client.get_insider_transactions(
        issuer_cik="320193",
        period_of_report_start_date="2025-01-01",
        period_of_report_end_date="2025-12-31",
        limit=750,
        cursor=cursor,
    )
    for tx in page["insiderTransactions"]:
        ...
    if not page["hasMore"]:
        break
    cursor = page["nextCursor"]

Each page is a separate request and counts against your monthly quota — prefer a large limit (up to 750) on big walks to minimize request count.

---

Usage examples

Recent Form 4 purchases by date range

result = client.get_insider_transactions(
    transaction_types="purchase",
    form_types="4",
    filing_date_in_est_start_date="2026-05-01",
    filing_date_in_est_end_date="2026-05-31",
    limit=50,
)

High-value transactions over $1M

result = client.get_insider_transactions(
    min_total_amount=1_000_000,
    transaction_types="purchase",
    filing_date_in_est_start_date="2026-05-01",
    filing_date_in_est_end_date="2026-05-31",
    limit=25,
)

All transactions by a specific insider

result = client.get_insider_transactions(reporting_owner_cik="1214128", limit=20)

CEO transactions across all companies

result = client.get_insider_transactions(
    officer_titles=["CEO"],
    form_types="4",
    filing_date_in_est_start_date="2026-05-01",
    filing_date_in_est_end_date="2026-05-31",
    limit=20,
)

Look up a specific filing

result = client.get_insider_transactions(accession_number="0001140361-26-023363")

---

Typing

The package ships with type hints (py.typed). Responses are plain dicts; TypedDict definitions describe their shape for editors and type checkers.

from insider_trades_api import (
    InsiderTransaction,
    InsiderTransactionsResponse,
    IssuerInformation,
    ReportingOwnerInformation,
    NonDerivativeTransaction,
    DerivativeTransaction,
    NonDerivativeHolding,
    DerivativeHolding,
    Footnote,
    OwnerSignature,
)

Note: the has* flags (hasPurchases, hasSales, …) are returned as the string "true", not a Python bool. Use truthiness checks (if tx.get("hasPurchases"):) rather than == True.

---

Error handling

Client-side validation failures raise ValueError before any network call. Non-success HTTP responses raise InsiderTradesAPIError, which carries the HTTP status.

from insider_trades_api import InsiderTradesAPI, InsiderTradesAPIError

try:
    result = client.get_insider_transactions(issuer_ticker="AAPL")
except InsiderTradesAPIError as err:
    print(err.status, err)  # e.g. 403 "InsiderTrades API error 403: Forbidden"

Status	Meaning
400	Invalid parameters (bad date format, invalid CIK, inverted date range)
403	Invalid or missing API key, or requested date range exceeds plan limit
429	Rate limit exceeded — retried automatically up to 3 times with backoff
504	Request timed out — retried automatically up to 3 times
500	Server error

The SDK validates date format (YYYY-MM-DD), numeric CIK, and date range order before making any network request, so many errors are caught immediately without a round trip.

---

Plans and limits

Plan	Monthly requests	Historical depth	Rate limit
Free	1,000	1 year	2 req/s
Pro	100,000	10 years	10 req/s

The SDK retries automatically on 429 and 504 responses (up to 3 attempts, respects the Retry-After header).

---

Changelog

[0.1.0] - 2026-06-05

• Initial release: get_insider_transactions() with full query parameter support
• Filters: issuer_ticker, issuer_cik, reporting_owner_cik, transaction_types, owner_roles, officer_titles, date ranges, min_total_amount, max_total_amount, fieldset
• Retry logic for 429 and 504 with backoff
• Client-side input validation for date format, CIK format, and date range order
• Full type hints (py.typed) including transaction arrays, holdings, and boolean flags
• Zero runtime dependencies (stdlib only)

---

Resources

• Docs: insidertrades.us/docs
• Pricing: insidertrades.us/pricing
• Dashboard: insidertrades.us/dashboard
• Support: insidertrades.us

---

Legal disclaimer

The data provided through this SDK is for informational purposes only and does not constitute investment advice, financial advice, trading advice, or any other type of advice. GoodTech LLC makes no representations as to the accuracy, completeness, or timeliness of the data. You are solely responsible for your use of the data. Past filings and transactions do not predict future performance. Always consult a qualified financial professional before making investment decisions.

---

Licensed under the Apache 2.0 License. Copyright 2026 GoodTech LLC.