Federal court opinion search via CourtListener. Search 10M+ opinions, docket records, and case history by keyword or company name.
Scored across 3 tools
get_case (by docket ID/name), get_company_cases (by company), and search_cases (filtered browse) target distinct query intents. However, get_case's 'case name search' mode overlaps somewhat with search_cases, which could cause occasional misselection.
All names are snake_case verb_noun (get_case, get_company_cases, search_cases), which is predictable and readable. The verb shifts between get_ and search_, but this reflects real semantic differences rather than inconsistency.
Three tools is on the thin side for a legal research surface; opinion retrieval, docket entry detail, and citation lookups are plausible sub-operations that have no dedicated tool. It is scoped but borderline minimal.
Core search and retrieval paths are covered, but the surface is entirely read-only with no way to fetch full opinion text separately, list courts/jurisdictions, or look up citations. These are notable gaps an agent doing legal research would hit.
Get detailed information about a specific court case/docket by docket ID or case name search.
Inputs: court, case_name, docket_id.
Find all federal court cases involving a specific company. Returns opinions, dockets, and case summaries.
Inputs: company (required), court, limit, date_after.
Search federal court opinions and cases via CourtListener / RECAP Archive. Filter by court, date, and case type.
Inputs: q (required), type, court, limit, +1 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Add one secure layer between your agents and this server.