BREAKING
pdf-brain: Fully Local PDF Search
How the Pipeline Works
1
Parse PDF & Markdown
↓
2
Run local Ollama model
↓
3
Store vectors in libSQL
↓
4
Search offline
0
dim
embedding size
0
$
per-query cost
0
formats indexed
LLM Enrichment & SKOS Taxonomy
MCP Layer for AI Assistants
Retrieval Layer
MCP
●
Exposes index via Model Context Protocol
●
Grounds assistant answers in your docs
Install Options
●
Standalone binary, no runtime
●
curl script or global npm
Toward Local-First Knowledge Tools
AI NEWS BLITZ
A new open-source tool called pdf-brain brings semantic search to your files, all offline.