LIVE · running (digest quality still being tuned)

Second Brain

The source of everything at slowloop — a personal knowledge base where AI conversations pile up and connect.

Second Brain

OVERVIEW

A personal knowledge base where AI conversations, development decisions, build struggles, and idea seeds all pile up in one place. Managed in Obsidian, synced via iCloud, and read and written directly by Claude Code — a living LLM wiki.

The core is a structure that separates source from distillation — `raw/` holds the original chats, notes, and scraps as immutable source, and the LLM picks out only what has signal to organize and connect under `wiki/`. Blog posts start here. The Hermes bot pipes Discord thoughts into it. myWiki is an attempt to turn this structure into a product. second is the hub everything at slowloop passes through.

QUICK FACTS

StatusRunning (digest being tuned)

TypePersonal knowledge base (Obsidian)

Structureraw (immutable) · wiki (LLM-distilled) · export (published)

SynciCloud

AI integrationClaude Code reads and writes directly; ops run as skills

Started2026-06-03

Key features

01 raw / wiki / export split

Source immutable, distillation LLM-owned, published output separate

02 Signal-based digestion

Not every chat — only what clears the signal threshold becomes a wiki page (digest skill)

03 Content-candidate tagging

Marks firsthand story/lesson/tip/method as blog and video material

04 Wiki as blog engine

Pulls published output into export/ for the slowloop site and blog

RELEASE NOTES

2026-06-24 Skill token optimization — All 36 skills assigned dedicated models (Haiku·Sonnet·Opus split)

2026-06-24 Index cache quality pass — 294 low-quality entries re-processed; 635 total entries now high-quality

2026-06-23 `index/` layer added — raw-cache moved to top-level `index/`; raw·wiki·index 3-layer structure established

2026-06-23 First `ingest-index` run — 341 non-project raw files indexed

2026-06-22 Todoist MCP integration — triage·plan-todo skills for task management and planning pipeline

Related — how it was made

Series →

Story I Built a Second Brain So My AI Chats Wouldn't Disappear What a month of AI chats left behind, and why I split raw files from wiki pages. Story Before Building the App, I Tried CLI and Obsidian First I fed 130 scattered chat histories into a CLI to validate the hypothesis in a single day. Method The Hard Part Wasn't Collecting, It Was Digesting How I set up Claude Code to run the second wiki: digest rules, pin, pipeline, and hooks. Story Storing Is Only Half the Job: I Built a Search Skill Too If the wiki has the answer, shouldn't it be able to find it when asked? Story My Planning AI and Coding AI Were Working in Silos The problem of context breaking between agents, and how I solved it with a second brain Method AI Wrote It, But It's My Story How raw conversations stored in a wiki become blog posts Story It Didn't Exist, So I Started Building It Myself Dozens of PKM apps, nothing fit, so after an OAuth headache I decided to build my own.

COMMENTS