Technical comparison

Insiders Trades Sigma vs the rest

Seven vendors sell insider-trades data. Only one covers eight regulators in a single schema, opens its scoring, and ships an MCP server for AI agents. Here are the numbers.

Feature matrix

Vendor	Markets	Free tier	Backtest	Scoring	Score explain	FX-norm.	MCP / AI	Pricing
Insiders Trades Sigma	8 · FR, US, DE, CH, UK, CA, IT, ES	Yes	Yes	Yes	Yes	Yes	Yes	Free · web only, 15 pages/day · Pro €19/mo (10k API req) · Quant €99/mo (100k req)
Quiver Quantitative	1 · US only	No	Partial	No	No	No	No	$10 to $1k+/mo · Congress trading is paid
OpenInsider	2 · US, CA	Yes	No	No	No	No	No	Free (no API · HTML scraping only)
WhaleWisdom	1 · US only (13F)	No	Partial	No	No	No	No	$50–$500/mo · 13F-centric, not Form 4
Fintel	4 · US + some intl	No	Partial	Partial	No	No	No	$30–$200/mo · API on top tiers
Form4 API (sec-api.io)	1 · US only	No	No	No	No	No	No	$50–$500/mo per endpoint

Pricing and feature columns sourced from public vendor sites as of May 2026. Directional comparison for marketing use, not a contractual claim about competitors.

Why us

Coverage

28 regulators, one schema

AMF, SEC, BaFin, SIX SER, RNS, SEDI, Consob, CNMV, AFM, FSMA, Oslo Børs, Nasdaq Helsinki, Nasdaq Stockholm, Nasdaq Copenhagen, ASX, FMA and Euronext Dublin, same Declaration object whether the trade was filed in Paris, Sydney or Dublin. OpenInsider has 2 markets. Everyone else: 1.

Alpha

Backtest-enriched signals

Every declaration carries T+30 / T+60 / T+90 / T+160 / T+365 / T+730 forward returns. Both insider-view (from transactionDate) and retail-view (from pubDate+1). Win-rate aggregates included.

Price

Free read tier

5,000 requests/day on the entire dataset, no paywall. Quiver charges $10+ to see Congress trades. WhaleWisdom $50+. We monetise the scoring product, not the raw data.

MCP-native

One config line plugs the dataset into Claude Desktop, Cursor, or Windsurf as a tool. Ask in natural language; the agent calls /api/v1 directly.

Currency-normalised aggregates

?currency=EUR converts USD/GBP/CHF/CAD amounts using point-in-time daily FX rates anchored on pubDate. Compare a Senator's buy and a German Vorstand purchase on the same scale.

Trust

Transparent scoring

/api/v1/scoring/explain/{amfId} decomposes every 0–100 score into 9 weighted factors with raw points and caps. No black box. Audit any signal yourself.

Anonymisation

CH / UK quirks handled

SIX SER aggregates trades by share class with redacted identity; RNS posts as company-level PDM. Our merge layer normalises both into per-insider rows where possible, and flags structural anonymity otherwise.

History

20 years of data

Earliest filings go back to 2006 across the 28 regulators (SEC Form 4 and AMF among the oldest). Enough sample size for genuine backtests, not 6-month curve fits.

Code examples

cURL

# Top US BUY signals last 30 days, score >= 70, with backtest returns
curl "https://insiders-trades.com/api/v1/declarations\
?market=us&direction=BUY&minScore=70&since=2026-04-15&include=backtest&limit=20" \
  -H "Authorization: Bearer sit_live_XXXXXXXX"

Python

import requests

r = requests.get(
    "https://insiders-trades.com/api/v1/declarations",
    params={
        "market": "us",
        "direction": "BUY",
        "minScore": 70,
        "currency": "USD",
        "include": "backtest",
        "limit": 50,
    },
    headers={"Authorization": "Bearer sit_live_XXXXXXXX"},
    timeout=15,
).json()

for d in r["items"]:
    print(d["company"]["name"], d["signal"]["score"], d["backtest"]["return90d"])

JavaScript

const res = await fetch(
  "https://insiders-trades.com/api/v1/declarations" +
  "?market=fr&minScore=60&cluster=true&limit=20",
  { headers: { Authorization: "Bearer sit_live_XXXXXXXX" } }
);
const { items } = await res.json();
console.log(items.map(d => `${d.company.name} · ${d.signal.score}`));

Claude tool-use (MCP)

// claude_desktop_config.json
{
  "mcpServers": {
    "insiders-trades": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://insiders-trades.com/api/mcp"],
      "env": { "API_KEY": "sit_live_XXXXXXXX" }
    }
  }
}
// Then in Claude: "Find the top 10 BUY signals in Germany this month, explain the highest score."

Use cases

Quant fund

Build cross-jurisdiction insider-cluster factors, FX-normalised. Backtest results inline on every declaration. Score-explain endpoint for factor-research.

/declarations · /backtest · /scoring/explain

Retail dashboard

Plug /signals for daily top picks. Display marketCode flags. Include T+90 historical win rate for honesty.

/signals · /companies/{slug} · /markets

News desk

Poll /declarations?since=YYYY-MM-DD every hour. Filter on minAmount + isCluster=true for newsworthy events. Build alerts off the OOS-tracker.

/declarations?since · /oos-performance

AI agent

Wire the MCP server into Claude. Users ask 'biggest insider buy this week in Italy', agent calls /declarations?market=it&direction=BUY directly.

MCP · /api/v1/* (all)

← Back to documentation

InsidersTradesSigma