archive-crawler

To crawl personal content archives, this Skill scans designated local drives, Dropbox, Backblaze, and Gmail takeouts to build manifests and triage high-signal personal content for ingestion. It requires an explicit archive path allow-list in gbrain.yml to enforce safety fences.

Yes, you can ingest content from a Gmail takeout or Dropbox dump safely. The Skill enforces safety fences using an explicit path allow-list in gbrain.yml, preventing over-scanning and ensuring sensitive data is not leaked during the triage process.

The best way to triage personal writing and ideas for brain pages is using a manifest-driven workflow. The Skill filters noise, inventories items across local drives, and queues high-signal personal content for review and ingestion.

The manifest-driven workflow tracks inventory and progress across cloud dumps by building a comprehensive list of archived items. It records reactions, filters out noise, and queues high-signal personal content for ingestion.

Crawling personal archives requires an explicit path allow-list in gbrain.yml to prevent accidental ingestion of sensitive content. This safety-first mechanism ensures the crawler only scans designated directories, avoiding over-scanning and data leakage.

Yes, you need an archive path allow-list to prevent over-scanning during content triage. The Skill mandates this configuration to enforce strict safety fences, ensuring only explicitly allowed local drives and cloud archives are scanned for ingestion.