What problem does it solve?
It removes the manual work of turning articles, URLs, and images into a living Obsidian knowledge base by automating ingestion, page creation/update, entity/concept extraction, and cross-referencing while keeping concurrency and consistency safe.
Core Features & Use Cases
- Source ingestion into a wiki vault: Read a file/URL/image, derive a stable source record under
.raw/, and produce corresponding wiki/sources entries.
- Automatic entity and concept page updates: Create or update wiki pages for people, organizations, products, repos, and significant ideas, including loggable linking changes.
- Cross-references, indexing, and visibility: Update indexes (
wiki/index.md), caches (wiki/hot.md), overview (wiki/overview.md), and append to wiki/log.md, then flag contradictions via > [!contradiction] callouts.
- Guardrails for correctness: Uses transport selection (CLI/MCP/filesystem), methodology-mode routing (LYT/PARA/Zettelkasten/generic), per-file advisory locking (
wiki-lock), and delta tracking via .raw/.manifest.json to avoid re-processing unchanged sources.
Quick Start
Tell the AI to ingest the URL https://example.com/article-123 into your Obsidian wiki so it creates and links the relevant source, entity, and concept pages and records the operation in wiki/log.md.