US Court Records Intel

Federal + state court records intelligence via CourtListener + RECAP - party monitoring, attorney tracking, litigation portfolios, recent-filings alerts, document lookup.

What does US Court Records Intel do?

It hits the CourtListener REST API and the RECAP archive to return actionable case intelligence rather than raw search dumps - per-party rollups, attorney case-load by year, corporate-entity litigation portfolios with settlement-ratio estimates, recent-filings alerts since a date, and metadata + optional plain text for documents already archived in RECAP.

AI / RAG / Agent

Federal and state court records prepared for legal AI agents and paralegal LLMs. RECAP document text is returned in clean plain-text ready to embed, and party / attorney / portfolio rollups give an agent the structured context it needs to answer "what has this entity been sued for" without scraping PACER. Compatible with LangChain, LlamaIndex, Pinecone, Weaviate, Chroma, and MCP-aware agent runtimes (Claude Desktop, GPT, custom).

from apify_client import ApifyClient
from langchain.schema import Document
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings

client = ApifyClient("APIFY_TOKEN")
run = client.actor("you/court-records-intel").call(run_input={
    "mode": "litigation_portfolio",
    "entity": "Acme Corp",
    "includeRecapText": True,
})

splitter = RecursiveCharacterTextSplitter(chunk_size=1500, chunk_overlap=150)
raw = [
    Document(
        page_content=doc["plain_text"],
        metadata={
            "case_name": item["case_name"],
            "court": item["court"],
            "docket_number": item["docket_number"],
            "filed": item["date_filed"],
            "judge": item.get("judge"),
            "doc_id": doc["recap_id"],
        },
    )
    for item in client.dataset(run["defaultDatasetId"]).iterate_items()
    for doc in item.get("documents", []) if doc.get("plain_text")
]

Chroma.from_documents(splitter.split_documents(raw), OpenAIEmbeddings(),
                       collection_name="litigation-rag")

Features

• Party monitoring - feed in a person or company; get every case they appear in with role, status, jurisdiction, judge, filing + termination dates.
• Litigation portfolio analysis - corporate-entity rollup: active vs closed counts, settlement-ratio estimate, average case duration, top jurisdictions, top judges, top opposing counsel.
• Attorney tracking - case load by year, derived practice areas, frequency of opposing counsel.
• Recent filings monitor - new dockets in specified courts / by specified parties / matching keywords since a date.
• RECAP document lookup - URLs + metadata + optional extracted plain text for documents already in the free RECAP archive.
• Patent-litigation crossover (schema-only stub) - reserved field for a future USPTO assignee cross-reference.

Use cases

• Law firms - competitive intel on opposing firms, conflict-check pre-screen, attorney book-of-business audits.
• Debt collection - find every active case where a target is a defendant.
• Insurance and M&A due diligence - litigation-portfolio rollups for a target entity in one shot.
• Litigation funders - filter by nature of suit + entity to surface fundable cases.
• Investigative journalists - track filing patterns, cluster by judge / firm / party.
• Competitive intel - monitor patent-litigation hot courts (E.D. Tex.) or securities-litigation hot courts (S.D.N.Y.).

FAQ

Q: Is this legal? A: Yes. Federal court dockets and documents are public records. The actor pulls primarily from RECAP (a free, donation-funded mirror of PACER documents) and CourtListener's open API. No PACER credentials are required and no PACER fees are incurred for documents already in RECAP.

Q: What's RECAP? A: RECAP is a free public archive of federal court documents hosted by the Free Law Project. When PACER users install the RECAP browser extension, every document they pay PACER to download is also uploaded to RECAP for free public access. RECAP now mirrors millions of dockets and tens of millions of documents from all 94 federal districts, 13 circuit courts, the bankruptcy courts, and SCOTUS. This actor reads RECAP via CourtListener's API so you get PACER coverage without PACER bills.

Q: Why might a run fail? A: The three most common failure modes are (1) CourtListener API rate-limiting on very wide party / keyword searches (the actor handles backoff), (2) a target case existing in PACER but not yet mirrored to RECAP - that document will return a "not in RECAP" sentinel rather than crash the run, and (3) ambiguous party names that match dozens of unrelated entities - tighten by court, date window, or nature-of-suit code.

Q: How fresh is the data? A: RECAP coverage of new filings depends on PACER users with the RECAP extension - hot dockets (high-profile cases, securities litigation, IP cases) are mirrored within hours, while obscure cases may take days or weeks. Docket metadata via CourtListener's PACER-RSS integration is typically same-day for participating courts.

Q: Can I schedule this daily or weekly? A: Yes. Daily cron is appropriate for recent_filings and attorney_tracking (new dockets land daily). Weekly is fine for litigation_portfolio and party_search (rollup reports). Apify Schedules + dedupe on docket_id gives clean deltas.

Q: Does it integrate with my CRM or data tooling? A: Yes. Apify webhooks POST every run to HubSpot, Salesforce, Pipedrive, Apollo, Slack, or any HTTP endpoint. Zapier, Make.com, and n8n templates available. Dataset exports as JSON / CSV / Excel / XML for litigation-management systems (LegalTracker, Mitratech, Tymetrix), conflict-check workflows, or BI tools.

Q: How does pricing work? A: PAY_PER_EVENT. You pay per case / docket record emitted, with surcharges on per-document document_lookup events and per-attorney / per-portfolio rollups. You only pay for what the actor actually emits.

Related Actors

• sec-edgar-intel - cross-reference securities-litigation defendants against EDGAR filings (10-K risk factors, 8-K disclosures, restatements) for full due-diligence context.
• uspto-patent-intel - pull the patent portfolios of parties in E.D. Tex. / D. Del. / N.D. Cal. patent-litigation cases to map the asserted IP.
• b2b-sales-triggers - convert new lawsuits, judgments, and litigation-portfolio events into outbound sales triggers for legal-tech and insurance teams.

Integrations

• Zapier - push to HubSpot/Salesforce/Pipedrive/Apollo
• Make.com - workflow automation
• n8n - self-hosted automation
• Apify webhooks - POST to your endpoint
• API + dataset export (JSON/CSV/Excel/XML)
• MCP / AI agents - call from Claude/GPT/LangChain

Modes

Mode	Required inputs	Emits
party_search	party_names	case records + per-party rollup
case_search	case_numbers or keywords or courts	case records
attorney_tracking	attorney_names	case records + attorney rollup
litigation_portfolio	party_names (corporate)	case records + portfolio rollup
recent_filings	optional courts, party_names, keywords	case records since date_from
document_lookup	case_numbers	one record per RECAP document

Input

See .actor/INPUT_SCHEMA.json. Sample - litigation portfolio for a corporate entity:

{
    "mode": "litigation_portfolio",
    "courtlistener_api_key": "YOUR_KEY",
    "party_names": ["Apple Inc."],
    "date_from": "2018-01-01",
    "include_documents": false,
    "max_results": 500
}

A free CourtListener API key raises the rate limit from ~60 req/min to ~5000/hr. Sign up at https://www.courtlistener.com/help/api/rest/.

Output

Sample output: ./.actor/sample-output.json — copy-paste-ready preview of real-looking records.

First record inline:

{
  "record_type": "case",
  "mode": "case_lookup",
  "source": "courtlistener",
  "scraped_at": "2026-05-13T22:01:14Z",
  "available": true,
  "reason": null,
  "case_id": 4488210,
  "case_name": "Aurora Signal Holdings, LLC v. Helion Compute, Inc.",
  "case_name_short": "Aurora Signal v. Helion Compute",
  "court": "United States District Court for the Northern District of California",
  "court_short_name": "N.D. Cal.",
  "jurisdiction": "federal",
  "case_number": "3:26-cv-01882",
  "docket_number": "3:26-cv-01882-VC",
  "nature_of_suit": "830 Patent",
  "cause": "35:271 Patent Infringement",
  "filing_date": "2026-03-18",
  "terminated_date": null,
  "status": "active",
  "assigned_to_judge": "Hon. Vince Chhabria",
  "referred_to_judge": "Hon. Sallie Kim (Magistrate)",
  "parties": [
    {
      "name": "Aurora Signal Holdings, LLC",
      "role": "plaintiff",
      "attorneys": [
        "Jessica Vega-Roth",
        "Marcus L. Tan"
      ]
    },
    {
      "name": "Helion Compute, Inc.",
      "role": "defendant",
      "attorneys": [
        "Priya Sundararajan",
        "Connor Whelan"
      ]
    }
  ],
  "attorneys": [
    {
      "name": "Jessica Vega-Roth",
      "firm": "Vega-Roth IP PLLC",
      "role": "for plaintiff"
    },
    {
      "name": "Marcus L. Tan",
      "firm": "Vega-Roth IP PLLC",
      "role": "for plaintiff"
    },
    {
      "name": "Priya Sundararajan",
      "firm": "Wilson Sonsini Goodrich & Rosati",
      "role": "for defendant"
    },
    {
      "name": "Connor Whelan",
      "firm": "Wilson Sonsini Goodrich & Rosati",
      "role": "for defendant"
    }
  ],
  "recent_docket_entries": [
    {
      "entry_number": 18,
      "date_filed": "2026-05-09",
      "description": "ORDER granting in part and denying in part defendant's motion to stay pending IPR. Signed by Judge Chhabria."
    },
    {
      "entry_number": 17,
      "date_filed": "2026-04-28",
      "description": "REPLY in support of motion to stay (Helion Compute, Inc.)"
    },
    {
      "entry_number": 16,
      "date_filed": "2026-04-21",
      "description": "OPPOSITION to motion to stay filed by Aurora Signal Holdings"
    },
    {
      "entry_number": 15,
      "date_filed": "2026-04-07",
      "description": "MOTION to stay pending Inter Partes Review (Helion Compute, Inc.)"
    }
  ],
  "tags": [
    "patent",
    "ipr",
    "stay_motion"
  ],
  "recap_documents": [
    {
      "doc_id": 31182144,
      "description": "Order on motion to stay",
      "url": "https://www.courtlistener.com/recap/3-26-cv-01882/order-stay.pdf",
      "page_count": 14,
      "filed_date": "2026-05-09",
      "text": null
    }
  ],
  "party_name": null,
  "total_cases": null,
  "active_cases": null,
  "closed_cases": null,
  "cases_as_plaintiff_count": null,
  "cases_as_defendant_count": null,
  "top_courts": null,
  "sample_cases": null,
  "attorney_name": null,
  "firm": null,
  "case_count_year": null,
  "top_practice_areas": null,
  "opposing_attorneys": null,
  "entity_name": null,
  "settlement_ratio_estimate": null,
  "avg_case_duration_days": null,
  "jurisdictions": null,
  "top_judges": null,
  "top_opposing_counsel": null,
  "patent_crossover_flag": null
}

Sample case record:

{
    "record_type": "case",
    "mode": "party_search",
    "source": "courtlistener",
    "case_id": 64902131,
    "case_name": "Apple Inc. v. Samsung Electronics Co., Ltd.",
    "court": "United States District Court for the Northern District of California",
    "court_short_name": "cand",
    "jurisdiction": "federal",
    "case_number": "5:11-cv-01846",
    "nature_of_suit": "830 Patent",
    "filing_date": "2011-04-15",
    "terminated_date": "2018-12-27",
    "status": "settled",
    "assigned_to_judge": "Lucy H. Koh",
    "parties": [
        {"name": "Apple Inc.", "role": "plaintiff", "attorneys": ["..."]},
        {"name": "Samsung Electronics Co., Ltd.", "role": "defendant", "attorneys": ["..."]}
    ],
    "tags": ["patent"],
    "available": true,
    "scraped_at": "2026-05-14T12:00:00Z"
}

Pricing

Pay-per-event:

Event	Price
case_record	$0.005
document_record	$0.003
document_text_charge	$0.010
intelligence_record	$0.015

A typical party_search for one company that returns 50 cases costs 50 * $0.005 + 1 * $0.015 = $0.265.

FAQ

Q: Do I need a CourtListener API key? A: Optional but strongly recommended. Anonymous use is rate-limited to ~60 req/min; with a free key you get ~5000/hr.

Q: Does this hit PACER directly? A: No - PACER is paywalled and out of scope. We use only what RECAP has already archived. Coverage is excellent for federal district / appellate, thinner for state and bankruptcy.

Q: How accurate is settlement_ratio_estimate? A: It's a heuristic - we infer settlement from termination text ("stipulation of dismissal", "settle"), trial verdict from "judgment" / "verdict", dismissal from "dismiss". Directionally useful, not authoritative.

Q: Can I get document text? A: Set include_document_text=true. Text is only available where CourtListener has already OCRed / extracted the PDF; we do not run OCR ourselves.

Support

Open an issue on the actor's GitHub or email via the Apify Store contact link. Include the run ID and input config.

Changelog

See ./CHANGELOG.md.

Found this useful?

If this actor saved you time or money, please consider leaving a quick review on the Apify Store. Reviews help other buyers find work that solves their problem and let me prioritize the features paying customers actually use. Leave a review: https://apify.com/seibs.co/court-records-intel#reviews