Import Export Turns a PDF Library Into Portable Memory

articletoolpdf-brainknowledge-managementlocal-firstaimemorypdf-pipelinevault

pdf-brain import/export maps Joel's PDF corpus into a portable memory seed for joelclaw's docs and Vault workflows.

This Dropbox archive is Joel’s full pdf-brain library export, shared from Slack after adding import/export support and publishing the install path: npm install -g pdf-brain.

The useful bit isn’t the tarball. It’s the shape: a personal research library becomes portable system state. Not “some files on a drive,” not “a database I hope still works,” but an artifact that can be exported, moved, rehydrated, indexed, and handed to another machine without turning the whole thing into a brittle little ceremony.

pdf-brain describes itself as a local PDF and Markdown knowledge base with semantic search, AI enrichment, and SKOS taxonomy support. That lines up cleanly with joelclaw: local-first documents, indexed memory, source material that can flow into the Obsidian Vault, and enough import/export plumbing that the library isn’t trapped in one runtime.

This is the part worth remembering: export is a product feature, but import/export together is infrastructure. Once the corpus can leave and come back, it can be tested, backed up, forked, replayed, and used as a seed for Typesense, Inngest, or whatever weird retrieval rig comes next.

Key Ideas

  • pdf-brain has import/export capabilities for a local PDF and Markdown knowledge base.
  • A full library export turns a reading corpus into portable memory, which is much more useful than a pile of untracked files.
  • Import/export makes the corpus replayable across machines, indexes, and workflows instead of trapping it inside one local database.
  • The package metadata points at semantic search, AI enrichment, SKOS, Ollama, libSQL, and MCP as adjacent system concepts.
  • For joelclaw, the pattern maps to durable document memory: export the corpus, import it elsewhere, index it, and keep the source material portable.