document-indexing

Document indexing extracts structured metadata like content type, topics, and tools from your documents using AI. It classifies content (blog, tutorial, guide, etc.), identifies key topics and technologies, and detects structural elements like code examples, enabling searchable and organized content.

The indexing process supports multiple content types: BLOG, TUTORIAL, GUIDE, REFERENCE, WHITEPAPER, CASE_STUDY, FAQ, CHANGELOG, MARKETING, and OTHER. Each document receives classification into one of these categories along with extracted topics and technical details.

Yes, batch asynchronous indexing processes multiple documents concurrently with configurable concurrency settings. This enables rapid indexing of large document datasets while allowing re-indexing of existing content.

Extracted metadata includes content_type, extracted_title, primary_topics, tools_technologies, has_code_examples, has_step_by_step_procedures, and has_narrative_structure. These fields generate DocumentMetadata records optimized for search and clustering.

Yes, the skill handles batch processing of hundreds of technical articles with configurable concurrency. It's designed to automatically tag large datasets with content types, topics, and structural indicators for precise searches and recommendations.

Yes, the generated DocumentMetadata records support clustering and targeted content recommendations. By tagging documents with topics, tools, and structural features, you enable discovery systems to surface relevant content to users.