Release Notes — Updates from 3.4.8 to 3.4.27
We are pleased to announce the immediate availability of OpenLink Structured Data Sniffer (OSDS) 3.4.27, a powerful browser extension for discovering metadata embedded in HTML documents, analyzing URL structures, and enabling seamless integration between your browser’s Document Object Model (DOM) and a variety of Large Language Models (LLMs). These capabilities make it possible to automatically generate structured data for upload to Data Spaces such as Knowledge Graphs (via SPARQL endpoints) or Filesystems (using the Create, Save, and Share pattern).
This productivity tool offers the following benefits:
- Structured Data Extraction — Effortlessly retrieve content from your browser’s DOM across conventional web pages and social platforms (X/Twitter, Threads, BlueSky, Mastodon, Reddit, LinkedIn, and others). This content can serve as the basis for customizable, automatically generated LLM prompts that transform raw information into structured data and knowledge, ready for upload to Data Spaces (Knowledge Graphs or filesystems).
- Structured Data Transformation — Seamlessly discover and convert metadata embedded in HTML documents into multiple formats, including CSV, JSON, JSON-LD, Markdown, and RDF-Turtle.
- Entity Discovery and Management — Automatically identify, tag, and hyperlink text within your browser to enrich navigation, enhance data integration, and progressively grow Knowledge Graphs.
- Flexible Identity & Authenticity — Write safely to target Data Spaces using a variety of authentication protocols, while retaining full control over your logged-in identity.
New Features & Fixes
- Fixed issues with annotation dropdown menus in LLM-based chatbot interfaces.
- Fixed metadata parsing for Plain Old Semantic HTML (POSH).
- Fixed login state session issues during SPARQL uploads and/or when generating SuperLink overlays (which also support read–write operations to Data Spaces).
- Enhanced JSON parsing and RDF transformation.