chore: sync vault to ~/cosmic-brain (Obsidian-registered path)
This commit is contained in:
Daniel
155
wiki/Wiki Map.canvas
Normal file
155
wiki/Wiki Map.canvas
Normal file
@@ -0,0 +1,155 @@
|
||||
{
|
||||
"nodes": [
|
||||
{
|
||||
"id": "label-meta",
|
||||
"type": "text",
|
||||
"text": "## Navigation",
|
||||
"x": -480,
|
||||
"y": -260,
|
||||
"width": 160,
|
||||
"height": 50,
|
||||
"color": "5"
|
||||
},
|
||||
{
|
||||
"id": "label-concepts",
|
||||
"type": "text",
|
||||
"text": "## Concepts",
|
||||
"x": 200,
|
||||
"y": -260,
|
||||
"width": 140,
|
||||
"height": 50,
|
||||
"color": "1"
|
||||
},
|
||||
{
|
||||
"id": "label-entities",
|
||||
"type": "text",
|
||||
"text": "## Entities",
|
||||
"x": 200,
|
||||
"y": 160,
|
||||
"width": 140,
|
||||
"height": 50,
|
||||
"color": "3"
|
||||
},
|
||||
{
|
||||
"id": "label-sources",
|
||||
"type": "text",
|
||||
"text": "## Sources",
|
||||
"x": -480,
|
||||
"y": 160,
|
||||
"width": 140,
|
||||
"height": 50,
|
||||
"color": "4"
|
||||
},
|
||||
{
|
||||
"id": "n-index",
|
||||
"type": "file",
|
||||
"file": "wiki/index.md",
|
||||
"x": -460,
|
||||
"y": -200,
|
||||
"width": 200,
|
||||
"height": 80,
|
||||
"color": "5"
|
||||
},
|
||||
{
|
||||
"id": "n-hot",
|
||||
"type": "file",
|
||||
"file": "wiki/hot.md",
|
||||
"x": -460,
|
||||
"y": -100,
|
||||
"width": 200,
|
||||
"height": 80,
|
||||
"color": "5"
|
||||
},
|
||||
{
|
||||
"id": "n-log",
|
||||
"type": "file",
|
||||
"file": "wiki/log.md",
|
||||
"x": -460,
|
||||
"y": 0,
|
||||
"width": 200,
|
||||
"height": 80,
|
||||
"color": "5"
|
||||
},
|
||||
{
|
||||
"id": "n-overview",
|
||||
"type": "file",
|
||||
"file": "wiki/overview.md",
|
||||
"x": -460,
|
||||
"y": 100,
|
||||
"width": 200,
|
||||
"height": 80,
|
||||
"color": "5"
|
||||
},
|
||||
{
|
||||
"id": "n-dashboard",
|
||||
"type": "file",
|
||||
"file": "wiki/meta/dashboard.md",
|
||||
"x": -230,
|
||||
"y": -50,
|
||||
"width": 200,
|
||||
"height": 80,
|
||||
"color": "5"
|
||||
},
|
||||
{
|
||||
"id": "n-llm-wiki",
|
||||
"type": "file",
|
||||
"file": "wiki/concepts/LLM Wiki Pattern.md",
|
||||
"x": 200,
|
||||
"y": -200,
|
||||
"width": 220,
|
||||
"height": 80,
|
||||
"color": "1"
|
||||
},
|
||||
{
|
||||
"id": "n-hot-cache",
|
||||
"type": "file",
|
||||
"file": "wiki/concepts/Hot Cache.md",
|
||||
"x": 200,
|
||||
"y": -110,
|
||||
"width": 220,
|
||||
"height": 80,
|
||||
"color": "1"
|
||||
},
|
||||
{
|
||||
"id": "n-compound",
|
||||
"type": "file",
|
||||
"file": "wiki/concepts/Compounding Knowledge.md",
|
||||
"x": 200,
|
||||
"y": -20,
|
||||
"width": 220,
|
||||
"height": 80,
|
||||
"color": "1"
|
||||
},
|
||||
{
|
||||
"id": "n-karpathy",
|
||||
"type": "file",
|
||||
"file": "wiki/entities/Andrej Karpathy.md",
|
||||
"x": 200,
|
||||
"y": 210,
|
||||
"width": 220,
|
||||
"height": 80,
|
||||
"color": "3"
|
||||
},
|
||||
{
|
||||
"id": "n-source",
|
||||
"type": "file",
|
||||
"file": "wiki/sources/Nate Herk LLM Wiki Transcript.md",
|
||||
"x": -460,
|
||||
"y": 210,
|
||||
"width": 220,
|
||||
"height": 80,
|
||||
"color": "4"
|
||||
}
|
||||
],
|
||||
"edges": [
|
||||
{"id": "e1", "fromNode": "n-index", "fromSide": "right", "toNode": "n-dashboard", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e2", "fromNode": "n-hot", "fromSide": "right", "toNode": "n-dashboard", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e3", "fromNode": "n-log", "fromSide": "right", "toNode": "n-dashboard", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e4", "fromNode": "n-overview", "fromSide": "right", "toNode": "n-dashboard", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e5", "fromNode": "n-dashboard", "fromSide": "right", "toNode": "n-llm-wiki", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e6", "fromNode": "n-dashboard", "fromSide": "right", "toNode": "n-hot-cache", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e7", "fromNode": "n-dashboard", "fromSide": "right", "toNode": "n-compound", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e8", "fromNode": "n-dashboard", "fromSide": "bottom", "toNode": "n-karpathy", "toSide": "left", "toEnd": "arrow"},
|
||||
{"id": "e9", "fromNode": "n-source", "fromSide": "right", "toNode": "n-dashboard", "toSide": "bottom", "toEnd": "arrow"}
|
||||
]
|
||||
}
|
||||
70
wiki/concepts/Compounding Knowledge.md
Normal file
70
wiki/concepts/Compounding Knowledge.md
Normal file
@@ -0,0 +1,70 @@
|
||||
---
|
||||
type: concept
|
||||
title: "Compounding Knowledge"
|
||||
complexity: basic
|
||||
domain: knowledge-management
|
||||
aliases:
|
||||
- "Knowledge Compounding"
|
||||
- "Persistent Synthesis"
|
||||
created: 2026-04-07
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- concept
|
||||
- knowledge-management
|
||||
status: mature
|
||||
related:
|
||||
- "[[LLM Wiki Pattern]]"
|
||||
- "[[Hot Cache]]"
|
||||
- "[[Andrej Karpathy]]"
|
||||
- "[[Nate Herk LLM Wiki Transcript]]"
|
||||
sources:
|
||||
- "[[wiki/sources/Nate Herk LLM Wiki Transcript]]"
|
||||
---
|
||||
|
||||
# Compounding Knowledge
|
||||
|
||||
The central insight behind the [[LLM Wiki Pattern]]: knowledge in a wiki compounds like interest in a bank. Every source added, every question answered, every analysis filed makes the wiki more valuable — not just by adding pages, but by enriching the connections between existing pages.
|
||||
|
||||
---
|
||||
|
||||
## Why Normal AI Chats Don't Compound
|
||||
|
||||
In a standard chat, knowledge is ephemeral. Each session starts fresh. Even if you upload the same documents repeatedly, the LLM re-derives the same insights from scratch. Nothing accumulates.
|
||||
|
||||
The same is true of most RAG systems: they index raw documents and retrieve chunks at query time. The retrieval gets the right fragments, but no synthesis is built up. Nothing is compiled. Ask the same complex question twice and you get the same assembly process twice.
|
||||
|
||||
---
|
||||
|
||||
## How Wiki Knowledge Compounds
|
||||
|
||||
When a new source arrives, the LLM doesn't just index it. It integrates it:
|
||||
- Updates entity pages with new information
|
||||
- Flags contradictions with existing claims
|
||||
- Strengthens or challenges the evolving synthesis
|
||||
- Adds cross-references from the new source to existing pages and back
|
||||
|
||||
The cross-references are already there next time. The contradictions have already been flagged. The synthesis already reflects everything that was read.
|
||||
|
||||
**The wiki is pre-compiled knowledge.** RAG re-compiles on every query.
|
||||
|
||||
---
|
||||
|
||||
## The Maintenance Problem
|
||||
|
||||
Wikis maintained by humans decay. The maintenance burden grows faster than the value — updating cross-references, keeping summaries current, noting when new data contradicts old claims. Humans abandon wikis because no one wants to do the bookkeeping.
|
||||
|
||||
LLMs don't get bored. They don't forget to update a cross-reference. The cost of maintenance is near zero. This is the practical reason the wiki pattern works: the entity that's best at the tedious maintenance work is the same entity that reads and writes the wiki.
|
||||
|
||||
---
|
||||
|
||||
## In Practice
|
||||
|
||||
One X user turned 383 scattered files and over 100 meeting transcripts into a compact wiki and dropped token usage by 95% when querying with Claude. The drop came from two sources: better navigation (index + hot cache vs. full document search) and pre-compiled synthesis (no re-deriving the same insights from scratch).
|
||||
|
||||
---
|
||||
|
||||
## Connections
|
||||
|
||||
See [[LLM Wiki Pattern]] for the full architecture.
|
||||
See [[Hot Cache]] for the session context mechanism.
|
||||
See [[Andrej Karpathy]] for the origin of this framing.
|
||||
95
wiki/concepts/Hot Cache.md
Normal file
95
wiki/concepts/Hot Cache.md
Normal file
@@ -0,0 +1,95 @@
|
||||
---
|
||||
type: concept
|
||||
title: "Hot Cache"
|
||||
complexity: basic
|
||||
domain: knowledge-management
|
||||
aliases:
|
||||
- "hot.md"
|
||||
- "Session Cache"
|
||||
- "Context Cache"
|
||||
created: 2026-04-07
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- concept
|
||||
- knowledge-management
|
||||
- context
|
||||
status: mature
|
||||
related:
|
||||
- "[[LLM Wiki Pattern]]"
|
||||
- "[[Compounding Knowledge]]"
|
||||
- "[[index]]"
|
||||
- "[[hot]]"
|
||||
sources:
|
||||
- "[[wiki/sources/Nate Herk LLM Wiki Transcript]]"
|
||||
---
|
||||
|
||||
# Hot Cache
|
||||
|
||||
A ~500-word summary of the most recent context in the wiki vault. Stored in `wiki/hot.md`. Updated at the end of every session and after every significant ingest or query.
|
||||
|
||||
The hot cache exists to answer one question: "where did we leave off?" A new session reads `hot.md` first. If the answer is there, it skips crawling the rest of the wiki.
|
||||
|
||||
---
|
||||
|
||||
## What It Stores
|
||||
|
||||
- What was most recently ingested or discussed
|
||||
- Key recent facts and takeaways
|
||||
- Pages recently created or updated
|
||||
- Active threads and open questions
|
||||
- What the user is currently focused on
|
||||
|
||||
---
|
||||
|
||||
## Format
|
||||
|
||||
```markdown
|
||||
---
|
||||
type: meta
|
||||
title: "Hot Cache"
|
||||
updated: YYYY-MM-DDTHH:MM:SS
|
||||
---
|
||||
|
||||
# Recent Context
|
||||
|
||||
## Last Updated
|
||||
YYYY-MM-DD — [what happened]
|
||||
|
||||
## Key Recent Facts
|
||||
- [Most important recent takeaway]
|
||||
- [Second]
|
||||
|
||||
## Recent Changes
|
||||
- Created: [[New Page 1]]
|
||||
- Updated: [[Existing Page]] (added section on X)
|
||||
- Flagged: Contradiction between [[Page A]] and [[Page B]]
|
||||
|
||||
## Active Threads
|
||||
- User is researching [topic]
|
||||
- Open question: [thing being investigated]
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Rules
|
||||
|
||||
- Keep it under 500 words. It is a cache, not a journal.
|
||||
- Overwrite it completely each time. Not append-only.
|
||||
- One file. Not split by date.
|
||||
- Updated after every ingest, significant query, and at the end of each session.
|
||||
|
||||
---
|
||||
|
||||
## Why It Matters
|
||||
|
||||
Without the hot cache, every session starts cold: read the index (1000 tokens), read several domain sub-indexes, read several individual pages. With the hot cache, the first 500 tokens often have everything needed.
|
||||
|
||||
Nate Herk reported that adding `hot.md` to his executive assistant vault reduced the token cost of session startup significantly compared to crawling multiple wiki pages.
|
||||
|
||||
The hot cache is especially valuable in cross-project setups: another Claude Code project can point at this vault and read `hot.md` first to get recent context at minimal token cost.
|
||||
|
||||
---
|
||||
|
||||
## Connections
|
||||
|
||||
The hot cache is part of the [[LLM Wiki Pattern]] token discipline strategy. See [[index]] for how the broader navigation works.
|
||||
98
wiki/concepts/LLM Wiki Pattern.md
Normal file
98
wiki/concepts/LLM Wiki Pattern.md
Normal file
@@ -0,0 +1,98 @@
|
||||
---
|
||||
type: concept
|
||||
title: "LLM Wiki Pattern"
|
||||
complexity: intermediate
|
||||
domain: knowledge-management
|
||||
aliases:
|
||||
- "LLM Knowledge Base"
|
||||
- "Karpathy Wiki"
|
||||
- "Persistent Wiki"
|
||||
created: 2026-04-07
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- concept
|
||||
- knowledge-management
|
||||
- llm
|
||||
- obsidian
|
||||
status: mature
|
||||
related:
|
||||
- "[[Hot Cache]]"
|
||||
- "[[Compounding Knowledge]]"
|
||||
- "[[Andrej Karpathy]]"
|
||||
- "[[Nate Herk LLM Wiki Transcript]]"
|
||||
- "[[index]]"
|
||||
sources:
|
||||
- "[[wiki/sources/Nate Herk LLM Wiki Transcript]]"
|
||||
---
|
||||
|
||||
# LLM Wiki Pattern
|
||||
|
||||
A pattern for building persistent, compounding knowledge bases using LLMs. Originated by [[Andrej Karpathy]]. The key insight: instead of re-deriving knowledge from raw documents on every query (RAG), the LLM incrementally builds and maintains a structured wiki that gets richer with every source added.
|
||||
|
||||
---
|
||||
|
||||
## The Core Idea
|
||||
|
||||
Most AI knowledge tools work like RAG: index raw documents, retrieve chunks at query time, generate an answer. Nothing accumulates. Ask a question that needs five documents and the LLM reassembles fragments every time.
|
||||
|
||||
The wiki pattern is different. When a new source arrives, the LLM reads it, extracts what matters, and integrates it into the wiki: updating entity pages, noting contradictions, strengthening the synthesis. The cross-references are already there. The knowledge is compiled once and kept current.
|
||||
|
||||
**The wiki is a persistent, compounding artifact.** The human curates sources and asks questions. The LLM writes and maintains everything.
|
||||
|
||||
---
|
||||
|
||||
## Three Layers
|
||||
|
||||
```
|
||||
.raw/ Layer 1 — immutable source documents
|
||||
wiki/ Layer 2 — LLM-generated knowledge base
|
||||
CLAUDE.md Layer 3 — schema that tells the LLM how to maintain it
|
||||
```
|
||||
|
||||
The LLM owns Layer 2 entirely. It creates pages, updates them when new sources arrive, maintains cross-references, and keeps everything consistent. The human reads; the LLM writes.
|
||||
|
||||
---
|
||||
|
||||
## Operations
|
||||
|
||||
**Ingest** — drop a source into `.raw/`, tell the LLM to process it. The LLM reads the source, discusses key takeaways, writes a summary page, updates entity and concept pages, and logs the operation. One source typically touches 8-15 wiki pages.
|
||||
|
||||
**Query** — ask a question. The LLM reads the index to find relevant pages, synthesizes an answer with citations. Good answers get filed back into the wiki.
|
||||
|
||||
**Lint** — periodic health check. Find orphan pages, dead links, stale claims, missing cross-references.
|
||||
|
||||
---
|
||||
|
||||
## Index and Log
|
||||
|
||||
**index.md** — content-oriented. A catalog of all pages with one-line summaries, organized by category. The LLM reads this first on every query to find relevant pages.
|
||||
|
||||
**log.md** — chronological. Append-only record of every ingest, query, and lint pass. Parseable: `grep "^## \[" log.md | head -10`
|
||||
|
||||
---
|
||||
|
||||
## Why It Works
|
||||
|
||||
The tedious part of maintaining a knowledge base is bookkeeping: updating cross-references, noting when new data contradicts old claims, keeping summaries current. Humans abandon wikis because the maintenance burden grows faster than the value. LLMs don't get bored. The wiki stays maintained because the cost of maintenance is near zero.
|
||||
|
||||
At small scale (~100 sources, ~hundreds of pages), the index file is sufficient. No vector database, no embeddings, no infrastructure. Just markdown files.
|
||||
|
||||
---
|
||||
|
||||
## Comparison to RAG
|
||||
|
||||
| Dimension | LLM Wiki | Semantic RAG |
|
||||
|-----------|----------|-------------|
|
||||
| Finding | Reads index, follows links | Similarity search over embeddings |
|
||||
| Infrastructure | Just markdown files | Embedding model + vector DB |
|
||||
| Cost | Tokens only | Ongoing compute + storage |
|
||||
| Maintenance | Run a lint | Re-embed when content changes |
|
||||
| Scale limit | Hundreds of pages | Millions of documents |
|
||||
|
||||
---
|
||||
|
||||
## Connections
|
||||
|
||||
See [[Compounding Knowledge]] for why the pattern produces more value over time.
|
||||
See [[Hot Cache]] for the session context optimization.
|
||||
See [[Andrej Karpathy]] for the pattern's origin.
|
||||
48
wiki/entities/Andrej Karpathy.md
Normal file
48
wiki/entities/Andrej Karpathy.md
Normal file
@@ -0,0 +1,48 @@
|
||||
---
|
||||
type: entity
|
||||
title: "Andrej Karpathy"
|
||||
entity_type: person
|
||||
role: "AI researcher, educator, founder"
|
||||
first_mentioned: "[[wiki/sources/Nate Herk LLM Wiki Transcript]]"
|
||||
created: 2026-04-07
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- entity
|
||||
- person
|
||||
- ai-researcher
|
||||
status: mature
|
||||
related:
|
||||
- "[[LLM Wiki Pattern]]"
|
||||
- "[[Compounding Knowledge]]"
|
||||
- "[[Nate Herk LLM Wiki Transcript]]"
|
||||
sources:
|
||||
- "[[wiki/sources/Nate Herk LLM Wiki Transcript]]"
|
||||
---
|
||||
|
||||
# Andrej Karpathy
|
||||
|
||||
AI researcher and educator. Former Director of AI at Tesla, founding member of OpenAI. Known for deep learning education (Neural Networks: Zero to Hero series, micrograd, nanoGPT) and for publishing practical AI patterns.
|
||||
|
||||
---
|
||||
|
||||
## Key Contributions to This Wiki
|
||||
|
||||
Karpathy originated the [[LLM Wiki Pattern]] — the idea of using LLMs to build and maintain a persistent, compounding knowledge base rather than re-deriving knowledge from raw documents on every query.
|
||||
|
||||
He published the idea as an "idea file" — intentionally abstract, designed to be copy-pasted into a Claude Code or similar session and built out collaboratively. The cosmic-brain plugin is a production implementation of this pattern.
|
||||
|
||||
His framing: "The wiki is a persistent, compounding artifact. The cross-references are already there. The contradictions have already been flagged. The synthesis already reflects everything you've read."
|
||||
|
||||
---
|
||||
|
||||
## Notable Quote
|
||||
|
||||
"I thought that I had to reach for fancy RAG, but the LLM has been pretty good at auto-maintaining index files and brief summaries of all documents and reads all the important related data fairly easily at this small scale."
|
||||
|
||||
---
|
||||
|
||||
## Connections
|
||||
|
||||
- [[LLM Wiki Pattern]] — his core contribution to knowledge management
|
||||
- [[Compounding Knowledge]] — the central insight he articulated
|
||||
- cosmic-brain plugin — this repo is a production implementation of his pattern
|
||||
38
wiki/hot.md
Normal file
38
wiki/hot.md
Normal file
@@ -0,0 +1,38 @@
|
||||
---
|
||||
type: meta
|
||||
title: "Hot Cache"
|
||||
updated: 2026-04-07T12:00:00
|
||||
tags:
|
||||
- meta
|
||||
- hot-cache
|
||||
status: evergreen
|
||||
related:
|
||||
- "[[index]]"
|
||||
- "[[log]]"
|
||||
---
|
||||
|
||||
# Recent Context
|
||||
|
||||
Navigation: [[index]] | [[log]] | [[overview]]
|
||||
|
||||
---
|
||||
|
||||
## Last Updated
|
||||
2026-04-07 — cosmic-brain v1.1.0 initialized. First ingest complete: Nate Herk LLM Wiki transcript.
|
||||
|
||||
## Key Recent Facts
|
||||
- The wiki pattern was originated by [[Andrej Karpathy]] and makes knowledge compound like interest
|
||||
- [[Hot Cache]] (this file) enables cross-project context at minimal token cost (~500 tokens)
|
||||
- [[Compounding Knowledge]] is the core insight: unlike RAG, the wiki pre-compiles synthesis
|
||||
- One user dropped token usage by 95% switching from inline context files to the wiki pattern
|
||||
- The index file alone is sufficient at hundreds of pages — no vector RAG needed
|
||||
|
||||
## Recent Changes
|
||||
- Created: [[LLM Wiki Pattern]], [[Hot Cache]], [[Compounding Knowledge]], [[Andrej Karpathy]]
|
||||
- Created: [[Nate Herk LLM Wiki Transcript]] (first source)
|
||||
- Created: seed vault structure (index, log, overview, dashboard)
|
||||
- Plugin v1.1.0: added /save and /autoresearch commands
|
||||
|
||||
## Active Threads
|
||||
- Vault is fresh. No user-specific domain scaffolded yet.
|
||||
- Next step: run /wiki to scaffold for your domain, then start ingesting sources.
|
||||
57
wiki/index.md
Normal file
57
wiki/index.md
Normal file
@@ -0,0 +1,57 @@
|
||||
---
|
||||
type: meta
|
||||
title: "Wiki Index"
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- meta
|
||||
- index
|
||||
status: evergreen
|
||||
related:
|
||||
- "[[overview]]"
|
||||
- "[[log]]"
|
||||
- "[[hot]]"
|
||||
---
|
||||
|
||||
# Wiki Index
|
||||
|
||||
Last updated: 2026-04-07 | Total pages: 11 | Sources ingested: 1
|
||||
|
||||
Navigation: [[overview]] | [[log]] | [[hot]] | [[dashboard]] | [[Wiki Map]]
|
||||
|
||||
---
|
||||
|
||||
## Concepts
|
||||
|
||||
- [[LLM Wiki Pattern]] — the pattern for building persistent, compounding knowledge bases using LLMs (status: mature)
|
||||
- [[Hot Cache]] — ~500-word session context file, updated after every ingest and session (status: mature)
|
||||
- [[Compounding Knowledge]] — why wiki knowledge grows more valuable over time, unlike RAG (status: mature)
|
||||
|
||||
---
|
||||
|
||||
## Entities
|
||||
|
||||
- [[Andrej Karpathy]] — AI researcher, originated the LLM wiki pattern (first: [[Nate Herk LLM Wiki Transcript]])
|
||||
|
||||
---
|
||||
|
||||
## Sources
|
||||
|
||||
- [[Nate Herk LLM Wiki Transcript]] — Nate Herk, 2026-04-07, transcript
|
||||
|
||||
---
|
||||
|
||||
## Questions
|
||||
|
||||
<!-- Add question entries here after queries or /save -->
|
||||
|
||||
---
|
||||
|
||||
## Comparisons
|
||||
|
||||
<!-- Add comparison entries here -->
|
||||
|
||||
---
|
||||
|
||||
## Domains
|
||||
|
||||
<!-- Add domain entries here after scaffold -->
|
||||
38
wiki/log.md
Normal file
38
wiki/log.md
Normal file
@@ -0,0 +1,38 @@
|
||||
---
|
||||
type: meta
|
||||
title: "Operation Log"
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- meta
|
||||
- log
|
||||
status: evergreen
|
||||
related:
|
||||
- "[[index]]"
|
||||
- "[[hot]]"
|
||||
---
|
||||
|
||||
# Operation Log
|
||||
|
||||
Navigation: [[index]] | [[hot]] | [[overview]]
|
||||
|
||||
Append-only. New entries go at the TOP. Never edit past entries.
|
||||
|
||||
Entry format: `## [YYYY-MM-DD] operation | Title`
|
||||
|
||||
Parse recent entries: `grep "^## \[" wiki/log.md | head -10`
|
||||
|
||||
---
|
||||
|
||||
## [2026-04-07] ingest | Nate Herk LLM Wiki Transcript
|
||||
|
||||
- Source: `.raw/nate-herk-llm-wiki-transcript.md`
|
||||
- Summary: [[Nate Herk LLM Wiki Transcript]]
|
||||
- Pages created: [[Nate Herk LLM Wiki Transcript]], [[LLM Wiki Pattern]], [[Hot Cache]], [[Compounding Knowledge]], [[Andrej Karpathy]]
|
||||
- Pages updated: [[index]], [[log]], [[hot]], [[overview]]
|
||||
- Key insight: The wiki pattern turns ephemeral AI chat into compounding knowledge — one user dropped token usage by 95%.
|
||||
|
||||
## [2026-04-07] setup | Vault initialized
|
||||
|
||||
- Plugin: cosmic-brain v1.1.0
|
||||
- Structure: seed files + first ingest complete
|
||||
- Skills: wiki, wiki-ingest, wiki-query, wiki-lint, save, autoresearch
|
||||
59
wiki/meta/dashboard.md
Normal file
59
wiki/meta/dashboard.md
Normal file
@@ -0,0 +1,59 @@
|
||||
---
|
||||
type: meta
|
||||
title: "Dashboard"
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- meta
|
||||
- dashboard
|
||||
status: evergreen
|
||||
related:
|
||||
- "[[index]]"
|
||||
- "[[overview]]"
|
||||
- "[[log]]"
|
||||
---
|
||||
|
||||
# Wiki Dashboard
|
||||
|
||||
Navigation: [[index]] | [[overview]] | [[log]] | [[hot]]
|
||||
|
||||
Requires the **Dataview** plugin: Settings > Community Plugins > Browse > "Dataview".
|
||||
|
||||
---
|
||||
|
||||
## Recent Activity
|
||||
|
||||
```dataview
|
||||
TABLE type, status, updated FROM "wiki" SORT updated DESC LIMIT 15
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Seed Pages (Need Development)
|
||||
|
||||
```dataview
|
||||
LIST FROM "wiki" WHERE status = "seed" SORT updated ASC
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Entities Missing Sources
|
||||
|
||||
```dataview
|
||||
LIST FROM "wiki/entities" WHERE !sources OR length(sources) = 0
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Open Questions
|
||||
|
||||
```dataview
|
||||
LIST FROM "wiki/questions" WHERE answer_quality = "draft" SORT created DESC
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Sources by Confidence
|
||||
|
||||
```dataview
|
||||
TABLE author, date_published, confidence FROM "wiki/sources" SORT confidence ASC
|
||||
```
|
||||
65
wiki/overview.md
Normal file
65
wiki/overview.md
Normal file
@@ -0,0 +1,65 @@
|
||||
---
|
||||
type: overview
|
||||
title: "Wiki Overview"
|
||||
created: 2026-04-07
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- meta
|
||||
- overview
|
||||
status: developing
|
||||
related:
|
||||
- "[[index]]"
|
||||
- "[[hot]]"
|
||||
- "[[log]]"
|
||||
- "[[dashboard]]"
|
||||
- "[[LLM Wiki Pattern]]"
|
||||
sources:
|
||||
- "[[wiki/sources/Nate Herk LLM Wiki Transcript]]"
|
||||
---
|
||||
|
||||
# Wiki Overview
|
||||
|
||||
Navigation: [[index]] | [[hot]] | [[log]] | [[dashboard]]
|
||||
|
||||
---
|
||||
|
||||
## Purpose
|
||||
|
||||
This is the cosmic-brain demo vault. It demonstrates the [[LLM Wiki Pattern]] — a system for building persistent, compounding knowledge bases using Claude and Obsidian.
|
||||
|
||||
Run `/wiki` to scaffold this vault for your own domain and replace this overview.
|
||||
|
||||
---
|
||||
|
||||
## Current Seed Content
|
||||
|
||||
This vault ships with a seeded first ingest based on Nate Herk's breakdown of the Karpathy LLM wiki pattern.
|
||||
|
||||
**Concepts seeded:**
|
||||
- [[LLM Wiki Pattern]] — the core architecture
|
||||
- [[Hot Cache]] — session context mechanism
|
||||
- [[Compounding Knowledge]] — why the pattern works
|
||||
|
||||
**Entities seeded:**
|
||||
- [[Andrej Karpathy]] — originated the pattern
|
||||
|
||||
**Sources seeded:**
|
||||
- [[Nate Herk LLM Wiki Transcript]] — practical demonstration of the pattern
|
||||
|
||||
---
|
||||
|
||||
## Current State
|
||||
|
||||
- Sources ingested: 1
|
||||
- Wiki pages: 11
|
||||
- Last activity: 2026-04-07 (cosmic-brain v1.1.0 initialized)
|
||||
|
||||
---
|
||||
|
||||
## Key Themes
|
||||
|
||||
**Knowledge compounds.** Unlike RAG, the wiki pre-compiles synthesis. Cross-references are already there. Contradictions are flagged. Every ingest enriches existing pages rather than adding isolated chunks.
|
||||
|
||||
**The hot cache is the force multiplier.** A ~500-word file captures recent context. New sessions start with full context at minimal token cost.
|
||||
|
||||
**Obsidian is the IDE, Claude is the programmer.** The graph view shows what's connected. The human curates sources and asks questions. Claude writes and maintains everything else.
|
||||
93
wiki/sources/Nate Herk LLM Wiki Transcript.md
Normal file
93
wiki/sources/Nate Herk LLM Wiki Transcript.md
Normal file
@@ -0,0 +1,93 @@
|
||||
---
|
||||
type: source
|
||||
title: "Nate Herk LLM Wiki Transcript"
|
||||
source_type: transcript
|
||||
author: "Nate Herk"
|
||||
date_published: 2026-04-07
|
||||
url: "https://youtube.com/@nateherk"
|
||||
confidence: high
|
||||
key_claims:
|
||||
- "LLM wiki makes knowledge compound like interest — nothing is re-derived on every query"
|
||||
- "Hot cache (~500 words) enables cross-project context without crawling the full wiki"
|
||||
- "One article can generate 15-25 wiki pages with full cross-references"
|
||||
- "One user dropped token usage by 95% switching from inline context files to wiki"
|
||||
- "Obsidian is the IDE, Claude is the programmer, the wiki is the codebase"
|
||||
- "Index file is enough at small scale (~100 sources) — no RAG infrastructure needed"
|
||||
created: 2026-04-07
|
||||
updated: 2026-04-07
|
||||
tags:
|
||||
- source
|
||||
- llm-wiki
|
||||
- obsidian
|
||||
- karpathy
|
||||
status: mature
|
||||
related:
|
||||
- "[[LLM Wiki Pattern]]"
|
||||
- "[[Hot Cache]]"
|
||||
- "[[Compounding Knowledge]]"
|
||||
- "[[Andrej Karpathy]]"
|
||||
- "[[index]]"
|
||||
sources:
|
||||
- "[[.raw/nate-herk-llm-wiki-transcript.md]]"
|
||||
---
|
||||
|
||||
# Nate Herk LLM Wiki Transcript
|
||||
|
||||
Raw source: [[.raw/nate-herk-llm-wiki-transcript.md]]
|
||||
|
||||
Nate Herk demonstrates the [[LLM Wiki Pattern]] in practice. He shows two live vaults: one for his YouTube transcript archive (36 videos) and one personal second brain. He breaks down Andrej Karpathy's original post and shows a 5-minute setup workflow.
|
||||
|
||||
---
|
||||
|
||||
## Key Takeaways
|
||||
|
||||
**The core insight**: normal AI chats are ephemeral. The wiki makes knowledge compound. Every source ingested, every question answered, every analysis filed — all of it stays and grows richer over time.
|
||||
|
||||
**The stack is simple**: Claude Code + Obsidian + a folder of markdown files. No vector databases, no embeddings, no infrastructure. Just files and Claude.
|
||||
|
||||
**The hot cache**: a ~500-word file (`wiki/hot.md`) that captures recent context. In an executive assistant setup, this prevented having to crawl dozens of wiki pages at the start of each session. See [[Hot Cache]].
|
||||
|
||||
**Cross-project referencing**: other Claude Code projects can read this vault by pointing at it in their CLAUDE.md. Nate's executive assistant reads from his herk-brain vault. Token usage dropped significantly compared to inline context files.
|
||||
|
||||
**At scale**: the index file alone is sufficient for hundreds of pages. Vector RAG only becomes necessary at millions of documents.
|
||||
|
||||
---
|
||||
|
||||
## Obsidian as IDE
|
||||
|
||||
Obsidian is just a markdown viewer with graph visualization. The graph view shows which pages are hubs (many connections) and which are orphans (none). Real-time — you can watch the wiki grow as Claude creates pages.
|
||||
|
||||
The key Obsidian features used:
|
||||
- Graph view — visualize the knowledge structure
|
||||
- Backlinks — follow connections between pages
|
||||
- Dataview — query pages by frontmatter
|
||||
- Web Clipper — send articles directly to `.raw/` from any browser
|
||||
|
||||
---
|
||||
|
||||
## Workflow Demonstrated
|
||||
|
||||
1. Install Obsidian, create a vault
|
||||
2. Paste Karpathy's LLM wiki idea into Claude Code
|
||||
3. Claude scaffolds the structure (raw/, wiki/, CLAUDE.md, index, log)
|
||||
4. Drop a source into `.raw/` using Web Clipper
|
||||
5. Tell Claude: "ingest this"
|
||||
6. Claude reads, creates 15-25 wiki pages, cross-references everything
|
||||
7. Query the wiki for insights
|
||||
|
||||
The ingest for one article (AI 2027) took 10 minutes and created 23 pages: 1 source, 6 people, 5 organizations, 1 AI systems page, multiple concepts, plus an analysis.
|
||||
|
||||
---
|
||||
|
||||
## Entities Mentioned
|
||||
|
||||
- [[Andrej Karpathy]] — originated the LLM wiki pattern
|
||||
- Nate Herk — demonstrated the pattern in this video
|
||||
|
||||
---
|
||||
|
||||
## Connections
|
||||
|
||||
See [[LLM Wiki Pattern]] for the full architecture.
|
||||
See [[Compounding Knowledge]] for the core insight on why this works.
|
||||
See [[Hot Cache]] for the session context mechanism.
|
||||
Reference in New Issue
Block a user