semantic-extraction

Agent-ready CLI for static page inspection, search triage, and browser handoff.

Cite Lens: An AI Tool for Detecting Out-of-Scope and Out-of-Context Citations; Conference paper accepted at TPDL 2025

Agent-readable semantic browser layer for AI agents — token-efficient web understanding, semantic snapshots, persistent sessions, and MCP/CLI/Python APIs.

This is the code for the basic HPO agent demo demonstrated at HDUG November 2025 as a taster of future agent-based data extraction possibilities which are already well on their way.

A new package that enables users to input textual descriptions of visual design, layout, or interface concepts and returns structured representations or annotations derived from the description. It le

Semantic Extraction & Data Unification

Metadata and semantic intelligence layer for MailPlus without duplicating raw mail.

Auditable document-intelligence pipeline: deterministic ingestion + schema-validated semantic extraction + reviewer workbench + opt-in Neo4j knowledge graph and Anthropic LLM entity extraction behind a human review gate.

Medical, legal, finance docs, LLM Wiki — semantic skeleton first, before GraphRAG or graphify. 30%+ more nodes & communities.

High-performance Python normalization engine for Agent-Commerce-OS. Handles semantic extraction, HTML parsing, and structured RAG output.

Text-to-Knowledge-Graph: Extract semantic relationships (SVO triples) from domain-specific text and build knowledge graphs with network analysis metrics.

A repository containing environmental imagery segmentation using several apis.

A local-first EPUB study companion that turns book sections into structured, LLM-ready Markdown packets