Case Law in the Same Answer as Your Own Files
CourtListener is the Free Law Project's research platform for United States court opinions, dockets, judge profiles and oral arguments, and it is free to reach. Free Law Project runs a hosted MCP server for it, and AI Intelligence Hub connects to MCP servers as a client, so an agent can cite published case law in the same answer as an organization's own material. CourtListener stays the source for the public record.
How it connects
The MCP server is hosted by Free Law Project, and the platform is the client. The connection is configured once with a name, the server URL, the transport and any HTTP headers needed, and the tool list is fetched when the connection binds rather than coded in. Streamable HTTP, Server-Sent Events and WebSocket are the transports carried.
Both forms apply. A workflow reaches one named tool through an MCP node with its arguments fixed at design time, which suits a scheduled watch on a docket. An agent is handed the whole tool list and chooses as it works, which suits a research question. Where a step needs something the server does not expose, the HTTP Request node calls the CourtListener API directly.
Because the material is public, there is no credential scoping to settle on that side. The part that does need care is the other half of the answer: when a question reaches both CourtListener and the organization's own library, retrieval from the library runs under the asking user's identity with their access list applied as a pre-filter, so a public citation never arrives attached to internal material the person could not open.
What you can do together
- Ask a question that reaches both the published opinions and your own case files, and get one answer citing each.
- Put a scheduled workflow on a docket so a filing is noticed when it appears rather than when someone checks.
- Have an agent gather the authority relevant to a matter as a starting point for a lawyer, with every citation traceable to the opinion it came from.
- Keep the public research and the internal record in one place without copying court documents into the library.
A scenario
- SetupAn administrator adds the CourtListener server to AI Intelligence Hub. No credential scoping is needed because the material is public.
- A questionAn attorney asks how a suppression argument like the one in a current matter has been treated in the circuit. The agent reads CourtListener and cites the opinions.
- Both halvesThe same answer includes the recorded interview from the office's own library that the argument turns on, retrieved under the attorney's identity.
- A watchA workflow checks a docket each morning and notifies the team when a filing appears.
- HandoverThe attorney keeps the citations with the matter, each one traceable to its source rather than paraphrased.
What stays where
CourtListener remains the source for the public record
Opinions and dockets are read from it and not copied into the library, so the citation points at the authoritative text.
Your own material stays under the library's access model
Retrieval from it runs as the asking user with their access list pre-filtered into the query, which is what keeps a public answer from carrying private material into it.
Processing runs where you deploy the platform
Shared or dedicated SaaS, your own cloud subscription, or your own servers. Reaching the hosted server needs a network path to it, which an air-gapped deployment does not have.
Products and solutions
Next step
See it on your own CourtListener instance. We will show the connection made, the data moving and the output, then size it for your deployment.
Request a demonstration or write to sales@vidizmo.ai
sales@vidizmo.ai · vidizmo.ai/integrations/catalog/courtlistener