Inquisita

Inquisita is an AI-native legal technology platform for document processing, search, and analysis. It combines a multi-modal ingestion pipeline, a clean queryable data model, and an open agent interface that lets firms connect their own AI assistants directly to their case data.

Key Features and Functions include:

Multi-Modal Ingestion
Inquisita processes PDFs, Word documents, video, audio, and images. Documents are extracted, chunked, and embedded automatically so every page, clip, and segment is searchable and analyzable through the same tools and workflows. Firms upload once and get structured, vector-indexed content without building custom pipelines.

SQL and Semantic Search
Agents and users query documents, passages, and analysis results through a clean SQL interface backed by pgvector. Keyword and semantic search sit side by side, and every row links back to the underlying source. Complex filters, joins, and full-text search are all first-class.

Structured Analysis Jobs
Firms run asynchronous analysis jobs over matters and collections. Jobs can classify, extract structured data, score relevance, or summarize, at either the document or passage level, using configurable LLM intelligence tiers for cost and quality tradeoffs. Results are stored in the same data model and queryable alongside the source content.

Collections and Enrichment
Collections are labeled, filterable views over documents or passages, built from SQL or semantic queries. Analysis results can be linked to collections without duplicating data, turning raw output into browsable, shareable review sets with scoring, classifications, and metadata.

Agent Interface (MCP)
Inquisita ships with a Model Context Protocol (MCP) server that makes the full platform usable by any MCP-compatible AI agent, including Claude, ChatGPT, and custom agents. A small, discoverable set of tools lets agents manage matters, ingest documents, query data, run analysis jobs, and build collections. Firms can adopt the Inquisita UI, bring their own agents, or use both side by side, without being locked into a single vendor's model or workflow.

Security and Trust
The platform is built on enterprise-grade infrastructure designed to meet the security and privacy expectations of modern law firms. Client documents are never used to train public models, all data is isolated and protected, and role-based access controls ensure that sensitive information remains secure within each firm.