Federal Court Dockets Scraper - PACER/RECAP API

Search the public CourtListener RECAP Archive for PACER-derived federal court docket evidence. Filter by keyword, case name, court, filing date, or nature of suit and export source-linked case names, courts, exact docket filing dates, and matching archived-document entries with text snippets and links.

No PACER login or CourtListener API token is required. The actor uses CourtListener's public RECAP search feed (CourtListener's REST search API now requires an account token).

First run and pricing

Raw empty input uses a deterministic caseName=Microsoft Corporation search and returns at most 3 docket items. At the live pay-per-event price, the maximum default charge is $0.0061:

• minimum actor-start event: $0.0001 (one event per GB, minimum one);
• each default-dataset item: $0.002.

maxResults is a total response cap. Apify maximum-total-charge limits are honored before results are fetched.

Output

Each found item has stable source and interpretation fields:

Field	Meaning
setupStatus	found for a matching docket
caseName	Source-reported case name
court, courtId	CourtListener court name and identifier
dateFiled	Docket filing date (resolved exactly from date-filtered searches)
dateFiledRange	Only present when the exact filing date could not be resolved within the request budget: YYYY-MM-DD/YYYY-MM-DD
latestEntryDate	Most recent filing date among the matching documents returned
documents	Matching RECAP documents: entry number, attachment number, filing date, text snippet, and link
matchingDocumentsCount	Number of matching documents returned with this item
moreDocumentsAvailable	true when the public feed was capped, so more matching documents (or dockets filed that day) may exist
docketId, docketUrl	CourtListener docket ID and direct docket page
sourceRetrievedAt	UTC retrieval timestamp
coverageNotice, interpretationNotice	Mandatory evidence limitations

Legacy fields (caseNumber, chapter, dateTerminated, natureOfSuit, cause, juryDemand, jurisdictionType, assignedTo, pacerCaseId, parties, attorneys, firms) are kept for schema compatibility but are empty: the anonymous public feed does not expose them. Open docketUrl for the full docket.

docketEntriesCount remains as a backward-compatible alias for the number of matching documents returned with the search item. It is not a complete docket-entry count.

Empty arrays are emitted as [], not null. A bounded search with no result returns an explicit setupStatus=no_match evidence item.

Input

{
  "caseName": "Microsoft Corporation",
  "dateFiledFrom": "2025-01-01",
  "maxResults": 3
}

Field	Description
query	Full-text RECAP query, such as a company, topic, or chapter phrase
court	CourtListener court ID, such as nysd, cacd, nysb, or deb
caseName	Case-name filter
dateFiledFrom	Earliest filing date, YYYY-MM-DD
dateFiledTo	Latest filing date, YYYY-MM-DD; omitted searches are bounded to today
natureOfSuit	Nature-of-suit filter
maxResults	Total paid item cap, 1-5000; default 3

If raw API input contains no search criteria, the actor uses the deterministic Microsoft Corporation case-name / 3-item fallback. Visible search fields remain blank by default, so a buyer's explicit query is never combined with a hidden example filter.

Example output

{
  "setupStatus": "found",
  "caseName": "Jones v. Microsoft Corporation",
  "court": "District Court, W.D. Arkansas",
  "courtId": "arwd",
  "dateFiled": "2026-10-26",
  "latestEntryDate": "2026-10-28",
  "matchingDocumentsCount": 7,
  "moreDocumentsAvailable": false,
  "documents": [
    {
      "documentNumber": 7,
      "entryNumber": 7,
      "dateFiled": "2026-10-28",
      "documentType": "PACER Document",
      "documentUrl": "https://www.courtlistener.com/docket/75019868/7/jones-v-microsoft-corporation/"
    }
  ],
  "docketId": 75019868,
  "docketUrl": "https://www.courtlistener.com/docket/75019868/jones-v-microsoft-corporation/",
  "source": "CourtListener RECAP archive (PACER-derived search evidence)"
}

Values can change as courts and CourtListener update records. Read the linked docket and documents before relying on a result.

Coverage and legal limits

RECAP is a crowd-sourced archive of PACER-derived records. It is not PACER itself and is not a complete or official federal-court search. Coverage varies by court and case; parties, attorneys, documents, dates, and classifications can be delayed, missing, corrected, or incomplete. moreDocumentsAvailable=true explicitly means the embedded documents array is bounded.

A matching name does not establish identity. A docket, allegation, filing, or party label does not prove liability, debt validity, misconduct, judgment, discharge, case outcome, current status, legal effect, or document authenticity. A no-match result does not prove no case exists. Verify the linked CourtListener record and PACER when a legal, credit, collections, compliance, or due-diligence decision requires complete or certified evidence.

Reliability

The actor retries transient request failures, response-body failures, HTTP 408, 429, and 5xx responses with bounded backoff. It reads CourtListener's public RECAP search feed (20 matching documents per request) and walks the docket filing-date range newest-first, splitting date windows so each docket gets an exact filing date. Requests are spaced and bounded per run. No authenticated calls are made.

Integrations

Download JSON, CSV, or Excel from the Apify dataset, schedule recurring searches, or connect results through webhooks, Zapier, Make, Google Sheets, and the Apify API.