Regulatory Enforcement Search & Monitor

Search official SEC litigation, ESMA sanctions, CNIL sanctions, and FDIC bank-failure records in one normalized dataset. Run it with no input to discover recent records, or add company and person names for repeatable compliance monitoring.

Independent, unofficial tool: This Actor is not affiliated with, sponsored by, or endorsed by the SEC, ESMA, CNIL, or FDIC. “Official” refers only to the public records used as sources.

The Actor is built for compliance research, third-party due diligence, investigations, vendor monitoring, and alerting workflows. Every row links back to official evidence. It does not produce a risk score or legal conclusion.

Start in seconds

Open the Input tab and click Start. The default settings are already small and safe, so an empty input works:

{}

To check specific names, provide only a watchlist:

{
  "entityQueries": ["Example Holdings Ltd", "Example National Bank"]
}

Results appear in the default dataset and can be downloaded as JSON, CSV, Excel, XML, or other Apify-supported formats. RSS and JSON feeds are also created by default.

Official sources

• US Securities and Exchange Commission litigation releases
• European Securities and Markets Authority sanctions register
• French CNIL sanctions tables
• US Federal Deposit Insurance Corporation bank failures

FDIC failures are institution-risk signals, not enforcement actions. CNIL's English tables frequently disclose only an organization type, so those rows support discovery but are not treated as named-entity matches.

What you can do

• Discover recent enforcement, sanction, and bank-failure records
• Check up to 25 organization or person names in one run
• Schedule repeat checks and return only new or changed evidence
• Send normalized records into review, CRM, alerting, or data pipelines
• Audit why a name matched and whether an official record changed
• Consume results through the Apify API, CSV export, RSS, or JSON Feed

Simple input

Most users only need these four fields:

• entityQueries — optional company or person names. Leave empty for recent-record discovery.
• sources — official sources to search. All four are selected by default.
• lookbackDays — history window from 1 to 365 days. Default: 365.
• maxResultsPerSource — maximum rows returned from each source. Default: 3.

The collapsed Advanced options section contains only two choices:

• outputMode — return all matches, only new and changed records, or only new records.
• useAiMatching — optionally review ambiguous name matches after exact matching has run.

Monitoring state is namespaced automatically. Valid RSS and JSON feeds are created automatically. API users can still override monitoringKey or set includeFeeds to false, but normal Console users do not need to configure either field.

AI matching is off by default and requires at least one entityQueries value. It evaluates no more than 20 ambiguous candidates per run and uses a fixed production confidence threshold. AI never creates or changes official facts.

Useful, traceable output

Each dataset row contains:

• official authority, jurisdiction, record ID, source URL, and document links
• event date, category, event type, stage, and status
• disclosed entity names and identifiers
• official or clearly labelled constructed summary
• explicitly stated monetary penalties when conservative extraction is possible
• watchlist match method, confidence, reason, and matched query
• stable record and content hashes with new, changed, unchanged, or unknown status

The default dataset view puts the most useful review fields first: date, title, entities, summary, authority, jurisdiction, category, event type, match method, change status, and official source.

Example row:

{
  "schemaVersion": "1.0",
  "recordId": "…",
  "source": "fdic-bank-failures",
  "sourceRecordId": "12345",
  "signalCategory": "institution-failure",
  "eventType": "bank-failure",
  "authority": "Federal Deposit Insurance Corporation",
  "jurisdiction": "United States",
  "entityNames": ["Example National Bank"],
  "title": "Example National Bank — bank failure",
  "eventDate": "2026-01-01T00:00:00.000Z",
  "sourceUrl": "https://www.fdic.gov/bank-failures/failed-bank-list",
  "matchedQueries": ["Example National Bank"],
  "matchMethod": "exact_entity",
  "matchConfidence": 1,
  "changeType": "new",
  "contentHash": "…"
}

Every row is validated before it is written. Invalid dates, insecure source URLs, malformed hashes, missing required fields, and unexpected fields cause the run to fail instead of silently producing a broken dataset.

Python API example

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_TOKEN")

run = client.actor(
    "thescrapelab/regulatory-enforcement-search-monitor"
).call(run_input={
    "entityQueries": ["Example Holdings Ltd"],
})

for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["eventDate"], item["title"], item["sourceUrl"])

Monitoring and feeds

The Actor remembers previously seen records for the same sources, names, and history window. Use newAndChanged for scheduled alerts or newOnly when only first appearances matter. Records withheld by an Apify spending limit are not marked as delivered, so a later run can still return them.

Each successful run provides:

• normalized dataset results and CSV export
• RUN_SUMMARY with source health, counts, warnings, and billing limits
• RSS_XML and JSON_FEED for the rows delivered by that run
• AI_EVALUATIONS when optional AI matching is enabled

Data handling and privacy

• The Actor runs with limited permissions and cannot access unrelated resources in your Apify account.
• Exact matching runs inside the Actor. With useAiMatching: false (the default), watchlist queries and record evidence are not sent to an AI provider.
• If optional AI matching is enabled, only a bounded ambiguous-candidate payload is sent to the Actor owner's configured OpenRouter or OpenAI account. That payload can include the watchlist query, disclosed official entity names, record title, authority, event date, and a short official excerpt.
• Apify stores the run input and outputs according to your platform storage and retention settings. Monitoring state contains stable record hashes and timestamps; never submit names or other data you are not authorized to process.

Cost and performance

The Actor uses direct HTTP requests rather than a browser and defaults to 128 MB RAM. A small default run normally completes in seconds; the five-minute timeout leaves headroom for temporary official-source delays and bounded retries without increasing the charge for runs that finish earlier.

Pricing is pay per event, so you pay for useful work rather than a subscription:

• $0.00020 per run start
• $0.00049 per official evidence record delivered to the dataset ($0.49 per 1,000 records)
• $0.00099 per ambiguous candidate evaluated with optional AI matching

A no-match watchlist run costs only the start event. The default input can return at most 12 records and therefore costs at most about $0.0061 with AI disabled. A 100-record run costs about $0.0492. The Actor respects Apify's maximum run charge and stops adding payable rows before that limit is exceeded.

The Apify Console displays the active price and maximum run charge before a paid run. Start with the default three rows per source, inspect relevance, and increase the limit only when needed. AI matching is off by default, capped at 20 candidates, and never requires users to supply an AI provider key.

Reliability and limitations

• Public sources can change, throttle, go offline, correct records, or limit history.
• A single source failure does not discard valid results from other sources; the run summary reports partial coverage.
• If every selected source fails, the run fails rather than returning a misleading successful empty result.
• CNIL type-only rows cannot match a named watchlist entry.
• ESMA may publish anonymized subjects, which are not treated as named matches.
• SEC page structure varies, so penalty extraction remains conservative.
• Entity matching supports evidence triage, not identity resolution.
• This Actor is not a complete global enforcement database and does not provide legal, compliance, investment, or safety advice.

Always review high-impact decisions against the official URLs supplied in the dataset.

Troubleshooting and support

• No watchlist results: use the exact disclosed legal name and widen the history window.
• No CNIL watchlist matches: run discovery mode because the English tables commonly omit entity names.
• Fewer rows than requested: open RUN_SUMMARY to review source coverage and spending-limit status.
• changeType is unknown: monitoring state was temporarily unavailable, but outputMode: all allowed evidence delivery to continue.
• AI mode fails immediately: the Actor owner must configure the provider secret; users never provide it in input.

For support, contact contact@thescrapelab.com and include the Apify run ID. Never send API keys or other secrets.