research-intelligence

Automate literature ingestion by routing arXiv and Semantic Scholar queries into a Supabase-backed research store. This pipeline classifies and embeds paper resources automatically, creating a searchable vector database for your research findings.

Yes, Zotero integrates directly into the ingestion pipeline to populate a Supabase vector store. You can then use RAG-enabled surfaces to augment prompts with relevant findings extracted from your synchronized Zotero library.

Yes, Supabase, vector embeddings, and edge functions are required to support the scope, screening, and presentation phases. These components provide the structured backend necessary to ingest, classify, and surface literature findings.

Track provenance using built-in audit trails that record prompt_versions, research_sessions, and paper_resources. This ensures complete lineage tracking from initial discovery and screening through to final evidence extraction.

Sync an Obsidian vault using interactive surfaces and tooling designed for knowledge graph mapping. This allows RAG-augmented findings from the research database to integrate directly with your existing notes.

Yes, you can orchestrate end-to-end workflows spanning discovery, screening, and evidence extraction. The pipeline surfaces ingested literature findings directly in Cowork or app interfaces for final presentation.