Tools

Open source infrastructure for the research projects.

"I built an external brain because context switching was destroying me."

Core Infrastructure

brain-mcp359K messages · 106K embedded · 256ms queries

The core cognitive infrastructure. 359K messages queryable in 256ms. Semantic search across 106K embedded messages. The external brain that makes everything else possible.

For anyone building personal knowledge infrastructure or AI memory systems.

youtube-transcription-pipeline32K videos · 15K transcripts · 4,142 hrs watched

Local-first video transcription. Whisper/Parakeet ML. Zero API costs at scale. Watch and forget → transcribe, embed, retrieve.

For heavy video consumers who want searchable knowledge from their watch history.

Frameworks & Tools

the-7-framework

Mental health dimensional framework. Alternative to DSM categories. Body-based dimensions instead of symptom checklists.

For clinicians, researchers, or anyone exploring alternatives to categorical diagnosis.

bottleneck-principle

Human sovereignty framework for AI systems. SHELET protocol. 100% human agency, AI in the loop.

For developers building AI systems that preserve human decision authority.

brain-canvas

Give any LLM its own display. Send JSON, get interactive UI. Visual interface for AI outputs.

For developers who want LLMs to render rich, interactive interfaces.

mcp-sefaria-server

MCP server for querying the Sefaria Jewish text library. Access Torah texts through AI.

For AI applications that need access to Jewish texts and commentaries.

The Brain MCP Architecture

359K raw messages (2023-2026)
         ↓
    Embedding pipeline
         ↓
106K vectors in LanceDB (32x smaller than DuckDB)
         ↓
    Semantic search (256ms)
         ↓
  MCP server interface
         ↓
   Claude Code / Any LLM

External memory for minds that can't hold it all. Context preservation across sessions. The cognitive prosthetic that makes poly possible.

What The Brain Tracks

Conversations

359K messages

YouTube

32K videos tracked

GitHub

132 repos · 1,427 commits

Principles

8 SEED principles

Embeddings

106K vectors

Timeline

2023-2026

YouTube Consumption Patterns

"YouTube Pipeline Saved Me 4 Hours a Day. My best teachers have often been YouTube tutorials, reviews and demos. I found myself spending more than five hours a day watching videos at 2–3× speed."

1littlecoder

1,256 videos

ADDitude

1,181 videos

TwoMinutePapers

947 videos

Vayimaen

823 videos

AIDailyBrief

792 videos

Living Lchaim

781 videos

6,152 unique channels tracked. AI tutorials, ADHD education, Torah content.

All repositories →