What problem does it solve?
Manually importing raw notes, Confluence pages, and other source materials into a curated, provenance-aware knowledge base wiki is time-consuming and error-prone, especially when handling batch processing, skip logic for protected content, and freshness metadata tracking.
Core Features & Use Cases
- Multi-format raw file conversion: Automatically converts VTT transcripts, EML emails, and HTML files to Markdown with embedded source references and extracted text, so they are compatible with the wiki importer.
- Safe batch processing: Partitions notes into manageable batches, supports parallel processing sessions, and uses atomic batch claiming to prevent conflicts when multiple sessions run at once.
- Confluence page ingestion: Fetches Confluence content, saves it with full provenance metadata (source URL, creation date, last modified date), and automatically assigns freshness dates to new and updated wiki pages.
- Integrated maintenance workflows: Triggers automatic freshness checks after ingestion to identify pages that need curation, and respects skip rules for already-ingested or protected notes to avoid duplicate work.
Quick Start
Use the wiki-ingest skill to import all new raw notes from your raw/ folder into the knowledge base wiki.
