What problem does it solve? Legal questions often get shallow or hallucinated AI answers with fabricated statutes and case citations. This Skill enforces a disciplined five-phase consultation workflow—fact gathering before legal research, multi-round RAG retrieval from curated legal knowledge bases, and a mandatory 41-item anti-hallucination checklist—so every cited statute and case is traceable to a real retrieval hit. ## Core Features & Use Cases - Five-Phase Consultation Workflow: User profiling, environment detection, scenario classification across eight legal domains (contracts, torts, family, labor, criminal, administrative, corporate, IP), tiered retrieval, and stepwise action plans. - Adaptive Retrieval Routing: Automatically detects IMA native, MCP-bridged, or standalone environments, selects the correct knowledge base identifiers, and falls back to web search only when RAG is unavailable. - Complexity-Based Routing: Classifies matters as simple, standard, or deep, scaling search rounds from 2 to 7+ with tiered reading (full text, abstract, title-only) and optional cross-validation of statute currency and regional judicial standards. - Anti-Hallucination Enforcement: Every statute citation carries a validity status label (current, repealed, amended, pending verification), and outputs pass a 41-item verification checklist before delivery. - Use Case: A user asks "my landlord won't return my deposit and the lease was verbal—can I sue?" The Skill gathers facts, retrieves the Civil Code provisions and similar cases from the knowledge base, compares facts against statutory elements, and outputs a structured opinion with tiered citations and a stepwise action plan. ## Quick Start Describe your legal situation in plain language, such as "my employer fired me citing performance but I never signed the evaluation policy—can I claim 2N compensation?", and the Skill will run its full consultation workflow.
