From single-agent to multi-agent: a comprehensive review of LLM-based legal agents
A review maps LLM legal-agent tasks, evaluation benchmarks and technical challenges across single-agent and multi-agent approaches, offering a starting point for legal-AI research design. Evidence reviewed from primary article excerpts.
Workflow level n/a Source: Peer-reviewed source Outcome: Not empirically evaluated here
Research applicability
Substantive primary survey supports bounded methodological relevance to legal research and evaluation: retrieval, legal text tasks and comparative benchmark selection. Recommend as a research-method overview, without endorsing underlying systems or professional legal deployment.
Getting started
Editorial, untested guidance: read the cited primary source and verify its limits before adapting the method. No candidate code, models or tools executed.
Maintenance
Only the cited snapshot was reviewed; present installation, maintained service availability and latest-version parity are unverified.
License
Check original licences, data rights and provider terms; source access is not reuse permission. No installation or rights audit performed.
Workflow & human oversight
Workflow level unassigned: The review covers multiple systems and task families rather than one demonstrated researcher-supervised workflow. Human direction, checkpoints and output validation vary across those systems and cannot be assigned collectively. Inspect a chosen underlying method and its original oversight protocol before assigning L1–L4. Human control and gaps: The review covers multiple systems and task families rather than one demonstrated researcher-supervised workflow. Human direction, checkpoints and output validation vary across those systems and cannot be assigned collectively. Inspect a chosen underlying method and its original oversight protocol before assigning L1–L4.
Workflow level is not an evidence score; higher is not better evidence.
Approach
Substantive primary survey supports bounded methodological relevance to legal research and evaluation: retrieval, legal text tasks and comparative benchmark selection. Recommend as a research-method overview, without endorsing underlying systems or professional legal deployment.
Outcomes (as reported by the originating team)
Author-reported findings or documented capabilities only; no independent reproduction or expert validation was performed.
Limitations
Retrieved article excerpts, not an exhaustive audit of all reviewed studies. Survey synthesis is not an original or independently reproduced performance result. No universal workflow level or cross-jurisdictional reliability is established.
Reproducibility
Source-text review with frozen provenance. No candidate software, models, data uploads or research outcomes were executed or reproduced.
Adoption requirements
Select a research task and review domain/data rights and human validation requirements; no adoption trial performed.
Comments
No comments yet.
Add a comment
Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.