llm-wiki

To build a persistent Obsidian knowledge base with an LLM, drop your source materials into the raw directory and run the ingest workflow. The system processes sources once, updates related entity and concept pages, and automatically expands cross-references to form a coherent knowledge graph.

Maintaining cross-references across LLM sessions requires a persistent wiki workflow that updates entity and concept pages during ingestion. This approach keeps cross-links current and maintains a running synthesis, ensuring that insights accumulate coherently over time without losing context between sessions.

Yes, you can ingest sources and update your Obsidian wiki using standard Python tooling. The workflow ingests sources into a dedicated directory, automatically updates related wiki pages, and expands cross-references without requiring any external dependencies beyond the standard Python environment.

No external dependencies are required to manage a second brain for long-running research in Obsidian. The system relies on standard Python tooling and Obsidian-compatible workflows, maintaining a searchable index and running synthesis to support long-form research and learning projects.

An LLM-driven wiki workflow handles incremental source ingestion by processing new materials, updating related entity and concept pages, and expanding cross-references. It maintains a running synthesis and searchable index, allowing accumulated insights to persist and support ongoing knowledge management tasks.

An Obsidian knowledge base loses context between LLM sessions without a persistent workflow to track cross-references and synthesis. Implementing an incremental ingestion process that updates entity pages and maintains a running log ensures insights accumulate coherently across all sessions.