skills.rest
stevengonsalvez
stevengonsalvezCOMMUNITY
Explore all skills by this author →
GitHub

add-karpathy-llm-wiki

Ingests diverse sources into a persistent, cross-referenced wiki for NanoClaw groups.

Updated Apr 10, 2026
Search Tags:#ingestion#wiki#knowledge-management#cross-references#nanoclaw#karpathy#llm-wiki
One-click install
npx skills add https://github.com/stevengonsalvez/nanoclaw-standalone --skill add-karpathy-llm-wiki-stevengonsalvez
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: add-karpathy-llm-wiki
Source: https://github.com/stevengonsalvez/nanoclaw-standalone/tree/main/.claude/skills/add-karpathy-llm-wiki
Command: npx skills add https://github.com/stevengonsalvez/nanoclaw-standalone --skill add-karpathy-llm-wiki-stevengonsalvez

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Groups using NanoClaw often lack a coherent, persistent, cross-referenced knowledge base. This skill helps teams create and maintain a wiki that evolves as sources are ingested, surfacing summaries, entities, and connections.

Core Features & Use Cases

  • Ingest sources individually into a wiki, generating page summaries, entity and concept pages, and cross-references while updating logs.
  • Maintain a three-layer structure (raw sources, wiki, and schema/CLAUDE.md) for disciplined ingestion, querying, and maintenance.
  • Use cases include onboarding new members, conducting long-form research, documenting project knowledge, and building a living knowledge base within a NanoClaw group.

Quick Start

Provide the group name and initial sources to ingest, and I will scaffold the wiki, configure the container skill, and guide you through the first ingestion.

Frequently Asked Questions about add-karpathy-llm-wiki

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a persistent wiki knowledge base from diverse sources?▼

To build a persistent wiki, you ingest diverse sources individually into a NanoClaw group, which converts them into cross-referenced summaries, entity pages, and concept pages while updating logs.

What is the three-layer structure used for wiki knowledge management?▼

The three-layer structure for wiki knowledge management consists of raw sources, the generated wiki pages, and a schema/CLAUDE.md layer, ensuring disciplined ingestion, querying, and maintenance.

How do I create cross-references and entity pages during document ingestion?▼

During document ingestion, the system applies per-file discipline to automatically generate entity and concept pages, build summaries, and establish cross-references between related content within the wiki.

Can I use this for onboarding new team members and long-form research?▼

Yes, you can use this for onboarding new members and long-form research, as it maintains a living knowledge base that evolves as new sources are ingested and cross-referenced.

Do I need a NanoClaw group to set up a cross-referenced wiki?▼

Yes, you need a NanoClaw group to set up the wiki, as the skill scaffolds the knowledge base, configures the container skill, and guides ingestion within that group environment.

What's the best way to maintain and query a knowledge base as sources grow?▼

The best way to maintain and query a growing knowledge base is to use the integrated container skills for per-file ingestion, linting, and querying, which continuously update indexes and logs.

Related Skills

Explore More Content & Communication Skills→
BartunekBartunek

add-karpathy-llm-wiki

Creates and maintains wiki knowledge bases for NanoClaw groups using LLM Wiki pattern with CLAUDE.md integration.

Community
Intermediate
sheawinklersheawinkler
78

llm-wiki

Build and maintain an interlinked Markdown knowledge base with YAML frontmatter and SCHEMA compliance.

Community
Advanced
thmtzthmtz
5

add-karpathy-llm-wiki

Ingest sources into a persistent wiki knowledge base within a NanoClaw group.

Community
Advanced
ce-dot-netce-dot-net
3

ACE Learning from Execution

Capture insights from task executions into a persistent knowledge base.

Official
Intermediate