data-ingest

To ingest chat exports and logs into Obsidian vault pages, use data-ingest to detect multiple source formats, cluster content by topic, and write domain pages following the vault schema with automatic index updates.

You can import JSON, Markdown, plain text, CSV/TSV, and HTML formats into a knowledge management vault. The skill detects these source formats and routes the content to appropriate domain pages.

Data ingest enforces deduplication when writing to Obsidian to maintain a clean knowledge graph. This ensures imported text data, logs, and transcripts do not create redundant domain pages.

You can ingest CSV and JSON data dumps into vault pages without external dependencies by default. The skill processes arbitrary text data and routes it to vault domains without requiring additional setup.

Domain-aware clustering for ingesting transcripts groups content by topic before writing to vault pages. This enables knowledge consolidation across multiple data sources while following the existing vault schema.

The best way to automate index updates when importing plain text to Obsidian is using a data ingest skill that writes domain pages following the vault schema and automatically updates the index during the import process.