Litco is built on the Matter Agent, an AI agent that takes on long-running, complex work. It operates across full-featured document review, legal research, and file management platforms, learning the record, the evidence, and the governing law as it goes, and it delivers finished work product, whatever the assignment demands.
LitKit · Document review
LitKit is Litco’s document review platform, reinvented around agentic AI. Your familiar e-discovery workflow meets an AI expert that lives in your document repository. Reviewers load their own production or the other side’s, assign batches, code documents, and produce with Bates stamps and a privilege log, with the Matter Agent alongside them the whole way, seeing every document they see. It runs the review, surfaces the documents that decide the case, builds a chronology, and pulls what they need for deposition or trial, learning the case as it goes.
Review & Tag on LitKit reads an entire document collection against criteria written in plain English, and the Matter Agent can launch a run for you. Every document gets a tag and a recorded rationale. Reviewers watch each call appear on the board as the run progresses, then check the finished run against a blind sample they code themselves.
LitSpace · Files & sharing
Lawyers get LitSpace with every matter—a file home for pleadings, productions, correspondence, and work product. Ask the Matter Agent for a document, and the Matter Agent finds the file and hands it to you in the chat. When you ask the Matter Agent to draft something, the Matter Agent saves the finished Word document in LitSpace, next to the record.
LitLex · Legal research
Case law, statutes, and regulations with the Matter Agent beside every result. Coverage includes federal and state court decisions, published and unpublished, updated daily.
The Matter Agent
The Matter Agent runs through everything above. Give the Matter Agent an objective, and the Matter Agent plans, searches, reads, drafts, and checks until the job is done or a decision needs a person. The Matter Agent writes and runs scripts in a sealed sandbox, produces real Word documents and Excel workbooks, and applies the law to the evidence in your record. Because the Matter Agent remembers the matter as your team works, every session starts with full knowledge of the case.
Give the Matter Agent an assignment in the matter chat. The Matter Agent plans the work and delivers a finished draft.
The Matter Agent reads the documents, drawing on your team’s tags, the results of automated review runs, and its own searches of the record. The Matter Agent researches the factual and legal questions and drafts a memo. Litco checks every quote against its source, matching record quotes to the source document and case quotes to the reporter text. Litco saves the finished draft to your matter files in whatever format you asked for.
Give the Matter Agent an assignment and come back to finished work. The Matter Agent plans the steps, works through them unattended, and delivers a draft with every citation already verified. Prefer to work alongside the Matter Agent? It stays with you while you review documents, run legal research, or draft in Word.
Have it pull the operative complaint, find the brief where you made that argument last spring, or file the memo you asked it to draft into your matter’s workspace. It uses its knowledge of those files when it reviews documents or runs legal research.
Language models invent citations and misread cases. Litco checks every quotation and characterization against the source.
Litco checks each draft against its sources in independent ways: matching quotations to source text, confirming pin cites against the reporter’s pagination, testing stated holdings against cited passages, and reading later treatment through two separate passes that flag a case as overruled only when both agree. Litco runs these checks in software outside the model. When any check fails, the Matter Agent repairs the draft or delivers it with the failure marked.
Litco tests the models it routes your work to, and publishes the results, including the failures.
LitigationBench is Litco’s benchmark for language models on litigation work. Every model runs the same 59-task battery twice, once on its own and once inside Litco with the safeguards running.
Litco onboards a few firms at a time.