For the first time, AI research agents can access the nation’s largest state trial court dataset across Claude, ChatGPT, and Trellis Chat.
Trellis announced the launch of the Trellis plugin for ChatGPT and Trellis Chat. These additions follow the recent release of the Trellis MCP Connector for Claude, which enables users to use Trellis directly within Claude. With these capabilities, users can now access Trellis’s more than 2.5 billion state trial court records through Claude, ChatGPT, and Trellis Chat.
This new capability combines agentic AI with the nation’s largest state trial court dataset, giving litigators a more powerful way to conduct research across the tools they already use.
Trial court intelligence, wherever you work
Attorneys can ask legal research questions directly within Claude or ChatGPT. The Trellis connectors enrich those conversations with relevant trial court data and litigation insights from Trellis, enabling attorneys to answer complex litigation questions with cited, jurisdiction-specific insights. Instead of relying solely on a model’s general knowledge, which does not include access to state trial court data, attorneys can tap into the intelligence available from the nation’s largest state trial court dataset without leaving the conversation. Trellis Chat delivers the same agentic experience natively within the Trellis platform.
The result was a memorandum, complete with visualizations and charts, that identified relevant California state trial court cases, surfaced the strongest factual analogues, and provided comparable verdicts as damages benchmarks, along with citations to the underlying filings and dockets. It was delivered in seconds instead of hours of manual research.
This goes beyond AI chat. Instead of simply answering a question, users can describe a research goal or desired work product. Across Claude, ChatGPT, and Trellis Chat, AI agents can use Trellis data to plan a research strategy, search billions of trial court records, cross-reference results against real verdicts, rulings, and motions, and verify citations before responding. These are the same steps a litigation associate would take when conducting legal research, compressed from hours into seconds.
Here’s what that looks like in practice. Recently, an AM100 partner asked Claude Cowork:
“Using Trellis, find factually similar California premises liability cases involving large retail stores and alleged notice of a spill hazard.”
The result was a memorandum, complete with visualizations and charts, that identified relevant California state trial court cases, surfaced the strongest factual analogues, and provided comparable verdicts as damages benchmarks, along with citations to the underlying filings and dockets. It was delivered in seconds instead of hours of manual research.
The data layer
State trial courts handle 98% of civil litigation in the United States, yet their records have historically been fragmented across thousands of jurisdictions and difficult to access at scale. While traditional legal research platforms help attorneys understand what the law says through appellate opinions, they provide little insight into how cases are actually litigated, how trial court judges tend to rule, what strategies succeed, or how opposing counsel approaches similar matters.
Trellis was built to close that gap. By aggregating and structuring billions of state trial court records into searchable litigation insights, Trellis enables attorneys to understand the people, patterns, and outcomes that shape litigation strategy, including judges, attorneys, experts, filings, and case outcomes.
“Agentic AI is changing how legal professionals conduct research. There is a real difference between a model that guesses at an answer from general memory and a system that does the research. It plans the work, searches the records, checks the outcomes, prepares the output in a Word document, double-checks its sources, and shows its work,” said Nicole Clark, co-founder and CEO of Trellis. “The U.S. state court system is the world’s largest, and combining agentic research with the nation’s largest state trial court dataset gives litigators a new superpower.”
Looking ahead
The Trellis MCP Connectors for Claude and ChatGPT are the first in a series of integrations Trellis plans to roll out, extending agentic access to its trial court data across the tools attorneys already use. Trellis Chat complements these integrations by providing the same experience natively within the Trellis platform.
“AI is quickly becoming a primary interface for legal work,” said Alon Shwartz, co-founder and COO of Trellis. “Our goal is to ensure attorneys can access Trellis’s litigation intelligence wherever they choose to work. The Trellis MCP Connectors for Claude and ChatGPT are just the beginning, with additional integrations planned to bring Trellis to even more of the platforms legal professionals rely on.”
With coverage spanning more than 3,500 courts across 2,500 counties in 45 states, Trellis is the leading source for state trial court data and litigation intelligence. Attorneys use Trellis to research judges, opposing counsel, experts, filings, rulings, verdicts, and other litigation insights that inform case strategy. As Trellis expands to more AI platforms, the goal remains the same: make state trial court intelligence available wherever legal professionals choose to work.
Access is subject to the Content View (CV) limits associated with an active Trellis subscription.
Contact sales@trellis.law for more details.