Litigon Groundwork Capability Sheet

A matter-grounded litigation research support worker that turns attorney asks, appellate signals, and case materials into authority-backed memos and threaded follow-up work.

Role

Litigation Research Support Worker

Mandate

Convert live litigation questions into grounded New York appellate research, usable memo output, and coherent follow-up work without losing the thread of the matter.

Supervision Model

Operates with attorney supervision on strategy, final legal judgment, and ambiguous scope shifts. Designed to preserve continuity and surface grounded authority, not to replace litigation judgment.

Owned Outcomes

  • Authority-backed research memos tied to active litigation questions
  • Retrieval and ranking of relevant New York appellate authority
  • Threaded follow-up handling across repeat asks on the same issue
  • Durable artifacts that can be reviewed, refined, and reused

Inputs

  • Plain-English research asks from an attorney
  • Matter facts, documents, and follow-up clarifications
  • Stored appellate corpus rows and prior generated artifacts
  • Email-threaded requests and replies

Systems Access

  • Local New York appellate corpus and citation store
  • Stored report and artifact history
  • Email-thread and conversation state
  • Queue-backed intake and processing layer

Outputs

  • Research memos
  • Short authority-backed email responses
  • Source lists and citations for verification
  • Durable context snapshots for follow-up work

Decision Rights

  • Retrieve and prioritize relevant appellate authorities
  • Draft first-pass research output from the maintained corpus
  • Preserve conversation continuity across follow-up requests
  • Request clarification when the ask is materially underspecified

Escalation Rules

  • Escalate when the issue requires litigation strategy rather than research support
  • Escalate when the facts are too incomplete to ground the answer safely
  • Escalate when the request moves outside the maintained New York appellate scope
  • Escalate when the follow-up becomes a new branch of work rather than a continuation

Reliability Notes

  • Strongest on narrow New York appellate research and repeat follow-up work
  • Reliability depends on maintained corpus quality and grounded citation paths
  • Not a substitute for attorney judgment, filing decisions, or client advice
  • Current live use is with an individual litigator rather than a full firm deployment

Litigon Groundwork should be understood as a litigation research support worker, not as a generic legal chatbot.

Its job is to take an attorney's live question, search a maintained New York appellate corpus, assemble grounded authorities, and return usable research output that can survive follow-up turns.

What the worker owns

The worker owns the research layer between a live litigation question and a usable first-pass answer:

  • retrieval
  • prioritization
  • memo assembly
  • continuity across follow-up turns
  • preservation of source grounding

That ownership is narrower than full legal strategy, but much more durable than one-off drafting.

It is also intentionally narrower than a whole-firm legal operating system. The current claim is about repeatable research support, not total workflow replacement.

Why this is more than prompting a model

A frontier model can produce a legal-looking answer from a prompt. That is not the same thing as a durable research support worker.

What makes Litigon Groundwork more operationally useful is the combination of:

  • a maintained New York appellate corpus instead of generic model recall
  • conversation and matter continuity across follow-up asks
  • stored artifacts and memo history that can be reviewed, refined, and reused
  • explicit supervision and escalation boundaries inside a real litigation workflow

The value is not only that it can draft text. The value is that it can return grounded research work product, keep the thread intact over time, and operate inside a repeatable support role rather than as an isolated one-shot answer engine.

Why the deployment matters

This is not only a speculative surface.

The system is already being used by a New York litigating attorney. That means the worker is exposed to real timing, real standards for usefulness, and real pressure to keep the thread of a matter coherent over time.

The current deployment is still user-first rather than firm-wide. That is a limitation, but it is also part of what makes the wedge legible. A durable worker can begin with one professional user before it becomes an organizational system.

What makes the worker credible

The stronger claim here is not broad autonomy.

It is that the worker can:

  • ground its output in maintained authority
  • produce reusable work product
  • preserve continuity across replies and follow-up asks
  • stay narrow enough that supervision and managed support remain explicit

That is what makes it operationally believable.