What problem does it solve? Answering factual or research questions usually means manually searching multiple sources, tracking citations, and losing the results afterward. This Skill automates multi-source research with Kagi and SearXNG, synthesizes findings with numbered citations, and persists every answer as a searchable note in an Obsidian vault. ## Core Features & Use Cases - Multi-Source Search & Synthesis: Queries Kagi tools (kagi_search, kagi_quick, kagi_assistant) plus SearXNG, expands queries into related angles, and runs a multi-pass synthesis (extraction, clustering, tension mapping, narrative). - Citation Tracking & Source Evaluation: Scripts track sources, assign sequential citation numbers, and score credibility by publication type and primary-source status. - Obsidian Persistence & Semantic Recall: Saves each answer as a dated Markdown note, embeds it via the NanoGPT embeddings API, and auto-ingests it into a cross-linked llm-wiki knowledge base. - Use Case: Ask "Compare Framework Laptop 16 vs ThinkPad P14s for Linux development" and receive an executive summary, key findings with citations, contradiction notes, and a saved research file linked to prior work. ## Quick Start Ask the agent to research the latest solid-state battery breakthroughs and save the cited findings to Obsidian.
