US Case Law API: opinions, dockets, courts and judges
Court opinions with full text, the dockets and judges behind them, citation resolution and a change feed. One REST API and MCP server, on the same key as the US statutes API.
Coverage per court, and how current each court is, is free to read at GET /api/v1/us/courts?include=coverage.
What the API returns
Opinions, the cases they belong to, the dockets they came from and the judges who wrote them, as one schema with stable ids.
Opinions with full text
GET /us/opinions/{opinionId}
The text of one opinion as HTML or plain text. A decision can carry a majority, concurrences and dissents, each its own opinion with author, per curiam flag and whether the text came from OCR. HTML marks every citation that points to a case we hold.
Cases
GET /us/cases/{caseId}
The decision itself: caption, court, decision date and its precision, publication status, every reporter citation, headmatter, syllabus, disposition, posture, counsel, the panel and the list of opinions.
Dockets
GET /us/dockets/{docketId}
Docket number, caption, key dates, assigned judges, nature of suit, appeal origin, the docket entries and the documents filed with each one. Oral arguments link back to their docket.
Courts
GET /us/courts
Every court with its readable id, system, level, state, parent court and active dates. Add include=coverage, also free, for what we hold per court and how current it is.
Judges
GET /us/judges/{judgeId}
Profile, positions held, how each was filled, dates. Case search can be scoped to a judge, and financial disclosure reports are listed per judge.
Citing and cited cases
GET /us/cases/{caseId}/citations
The cases a decision cites (direction=cites) or the cases that cite it (direction=cited_by, the default), each edge with a mention count. Resolve a list of reporter citations to cases in one call.
Change feed
GET /us/cases/changes
The cases added or updated since a moment, oldest first, in pages you follow with a cursor. Keep your own copy current without re-reading the corpus.
One request, cited decisions.
curl -X POST https://api.vaquill.ai/api/v1/us/cases/search \
-H "Authorization: Bearer $VAQUILL_API_KEY" \
-H "Content-Type: application/json" \
-d '{"query": "\"qualified immunity\" excessive force", "courtIds": ["US_FED_SCOTUS"], "limit": 1}'{
"results": [
{
"case": {
"caseId": "USCASE_3F9A2B1C4D5E6F708192",
"caseName": "Graham v. Connor",
"court": { "courtId": "US_FED_SCOTUS", "name": "Supreme Court of the United States" },
"decisionDate": "1989-05-15",
"decisionDatePrecision": "day",
"publicationStatus": "published",
"citations": ["490 U.S. 386"]
},
"opinionId": "USOP_3F9A2B1C4D5E6F708192"
}
],
"count": 1,
"limit": 1,
"offset": 0,
"hasMore": true,
"creditsConsumed": 4
}Every word must match; put a phrase in double quotes. Filter by court, state, date range, publication status or judge. Page with offset.
Resolve, read, follow
Turn a citation into a case, read the opinion behind it, and keep a copy current.
Resolve citations
Up to 100 reporter citations per call, one result each, in order. Each is resolved, ambiguous, not_found or unparsed.
curl -X POST https://api.vaquill.ai/api/v1/us/cases/resolve \
-H "Authorization: Bearer $VAQUILL_API_KEY" \
-H "Content-Type: application/json" \
-d '{"citations": ["410 U.S. 113"]}'Read an opinion
format=html keeps citation markup, format=text returns plain text. The response carries the opinion kind, author and a body.
curl "https://api.vaquill.ai/api/v1/us/opinions/USOP_3F9A2B1C4D5E6F708192?format=text" \
-H "Authorization: Bearer $VAQUILL_API_KEY"Follow the change feed
Pass the time of your last sync as since. Rows come oldest first; when hasMore is true, send nextCursor back as cursor.
curl "https://api.vaquill.ai/api/v1/us/cases/changes?since=2026-11-01T00:00:00Z&limit=2" \
-H "Authorization: Bearer $VAQUILL_API_KEY"{
"results": [
{
"caseId": "USCASE_3F9A2B1C4D5E6F708192",
"changeType": "upsert",
"change": "added",
"updatedAt": "2026-11-02T08:14:00Z",
"courtId": "US_FED_SCOTUS",
"caseName": "Graham v. Connor",
"decisionDate": "1989-05-15"
}
],
"count": 1,
"since": "2026-11-01T00:00:00Z",
"hasMore": true,
"nextCursor": "<opaque cursor>",
"creditsConsumed": 1
}Who it is for
- Developers
- Plain REST, JSON in camelCase, stable ids for every case, opinion, docket, court and judge, and the same Bearer key as the statutes API.
- Legal-tech builders
- Ground a research, citation-check or analytics product in the decisions themselves. Resolve citations in a brief, read the opinion behind each one, and show whether later courts treated it well.
- AI agents
- The MCP server carries case tools beside the statute tools: search, get a case, read an opinion, resolve citations and read up to 50 cases at once. An agent can move from a statute section to the cases that cite it.
Billing and bulk access
Metered in credits, per call, from the same balance as the rest of the Data API.
- A missing record and a server error are refunded. An empty page of results is an answer and is charged. A request that fails validation is never charged.
- Coverage reads are free. Per-endpoint prices are public at GET /api/v1/api-credits/pricing, which needs no key.
- Read up to 50 cases per call with POST /us/cases/batch. Resolve up to 100 citations per call.
Looking for statutes, regulations and court rules? That is the US statutes API, on the same key.
Frequently asked questions
Cases, the text of each opinion, dockets with their entries, courts, judges, citation resolution, the cases a decision cites and the cases citing it, and a change feed. It covers US courts only.