Court Listener logo

Court Listener

Search U.S. case law, opinions, dockets, filings, and patent references through CourtListener over MCP.

Install Court Listener

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "court-listener": {
      "command": "npx",
      "args": [
        "-y",
        "@pipeworx/mcp-court-listener"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Automated check passed— started and listed 11 tools correctly (8d ago).

Overview

Court Listener MCP server connects AI agents to Free Law Project’s CourtListener data for U.S. opinions, case records, RECAP dockets, and court filings. It supports name- and citation-based case lookup, full opinion retrieval, docket and filing searches, and patent references extracted from litigation records. Basic case and opinion access requires no authentication; upstream searches are throttled and can use a Free Law Project token. Reach for it when an agent needs structured legal research, especially across state-court opinions or litigation documents.

Use cases

  • Find a case by name or reporter citation
  • Retrieve majority opinions, dissents, and full text
  • Trace a docket from search through individual filings
  • Cross-reference patent litigation with USPTO metadata

Key features

  • Name- and citation-based case lookup
  • Full opinion text retrieval
  • Opinion and docket search
  • Docket sheet and filing access
  • Court filing text search
  • Patent references from RECAP litigation records

Tools11

What Court Listener lets your agent do. Expand a tool for its parameters.

Verified live~2.8k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

12/12 checks healthy over the last 64d

What it can do

From the 11 tools this server returned when we ran it; 11 declare their own safety hints.

Handles credentials · 7Makes network requests · 11Read-only tools · 11
Network at startup4 connections
Files written at startupNone
Install scriptsNone
npm provenanceSigned build
Package age129 days
Maintainers1

Observed 2026-11-02 in an isolated sandbox. A description of reach, not a security audit.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 1h ago via OSV.dev · @pipeworx/mcp-court-listener (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

npm downloads1kPackage downloads in the last 30 days.
Last commit1d agoMost recent push to the default branch.
Availability100%Our rolling endpoint + install checks that succeeded.
Install checkPassedOur sandbox started it and listed its tools.
Tools exposed11Callable tools this server registers over MCP.

How Court Listener works

What Court Listener does

Court Listener MCP server exposes selected CourtListener data and search paths as MCP tools. The corpus contains U.S. opinions from state, federal, territorial, military, and tribal courts, with state opinions making up most of the collection. It includes case names, reporter citations, judges, precedential status, citation counts, opinion text, and RECAP docket records.

The server separates judicial opinions from litigation filings. Opinions represent what a court decided; filings represent documents submitted by parties or the court, such as motions, briefs, and orders. That distinction matters when an agent needs either the holding in a case or the arguments and procedural history in the docket.

Tools and capabilities

  • find_case resolves a case by name and ranks matching results by citation count.
  • lookup_citation finds a case from a reporter citation such as 410 U.S. 113.
  • list_case_opinions lists the opinions associated with a case, including majority opinions, concurrences, and dissents.
  • get_opinion retrieves the full text for an opinion ID.
  • search_opinions searches by name, judge, category, or a specific court code. It may return a lead-opinion excerpt with the most-cited result.
  • search_dockets searches docket-level records upstream.
  • list_docket_filings returns the docket sheet for a docket_id.
  • get_docket_filing retrieves an individual filing and its PDF when one is available.
  • search_court_filings searches filing text rather than opinion text.
  • find_case_patents connects a RECAP docket to asserted patent numbers, defendants from the case caption, and optional USPTO metadata.

A typical opinion workflow is find_case, then list_case_opinions, then get_opinion. For litigation documents, use search_dockets, followed by list_docket_filings and get_docket_filing.

Setup and configuration

Basic case lookup and opinion-text access work without a key, signup, or authentication. Topic-based opinion searches and docket searches use upstream CourtListener search and can be subject to throttling; a free token from Free Law Project can lift that restriction.

find_case_patents can return asserted patent numbers without additional configuration. Setting PLATFORM_USPTO_KEY adds USPTO title, status, grant-date, and first-inventor metadata to those numbers. The key is the same USPTO Open Data Portal platform key used by the related patents pack.

Limitations and notes

Filing tools require a docket_id produced by search_dockets; a PACER case number does not resolve directly. RECAP contains many docket entries without PDFs because documents are only present when someone has obtained them from PACER. A missing PDF therefore indicates unavailable document content, not that the filing did not occur.

Filing searches share an upstream budget of roughly 125 requests per day without a token. Opinion tools are unaffected because opinion data can use a mirror. Responses include a snapshot_date, and opinions newer than that snapshot may require upstream search.

About 14.2% of the opinion corpus comes from OCR, and responses identify OCR-derived text with extracted_by_ocr. date_created records when CourtListener received a record, while date_filed indicates when the court issued the opinion. Citation counts are limited to the CourtListener corpus.

Patent extraction is intentionally conservative: numbers are treated as asserted patents only when the source text uses a literal patent-number label. Results may omit patents mentioned only in an unavailable complaint PDF. Defendants come from the indexed case caption, and an et al. caption is marked as truncated. Relaxed case-name matches should be checked against the returned case name before use.

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Court Listener

Court Listener is an MCP server that connects AI agents to Free Law Project’s CourtListener API for U.S. case law and RECAP filings. Its main tools find cases, resolve citations, retrieve opinions, search dockets, inspect filings, and cross-reference patent cases with USPTO metadata.

Installation instructions are not included here. The server exposes CourtListener capabilities over MCP and requires no authentication for basic case lookup and opinion text.

No, basic case lookup and opinion text work without authentication. A Free Law Project token can lift throttling for upstream opinion and docket searches, while `PLATFORM_USPTO_KEY` is optional for USPTO metadata in patent cross-references.

Search for a docket first, then pass its `docket_id` to `list_docket_filings` and use the resulting `document_id` with `get_docket_filing`. A PACER case number cannot be used directly, and some filings have no PDF available.

Yes, `find_case_patents` extracts asserted patent numbers from indexed RECAP text and identifies defendants from the case caption. With `PLATFORM_USPTO_KEY`, the results can also include USPTO titles, statuses, grant dates, and first inventors.

Related MCP servers

Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewCourt Listener AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/court-listener?style=directory)](https://allmcps.com/mcp/court-listener)
HTML Embed
<a href="https://allmcps.com/mcp/court-listener"><img src="https://allmcps.com/api/badge/court-listener?style=directory" alt="Court Listener on AllMCPs" /></a>