skills.rest
JamesFincher
JamesFincherCOMMUNITY
Explore all skills by this author →
GitHub

llm-wiki

Build an interlinked markdown knowledge base from URLs, PDFs, and pastes.

Updated May 4, 2026
Search Tags:#linting#markdown#rag#provenance#wiki#knowledge-base#entity-creation
One-click install
npx skills add https://github.com/JamesFincher/gengar --skill llm-wiki-jamesfincher
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: llm-wiki
Source: https://github.com/JamesFincher/gengar/tree/main/skills/research/llm-wiki
Command: npx skills add https://github.com/JamesFincher/gengar --skill llm-wiki-jamesfincher

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of scattered notes and repeated research by turning sources into a persistent, interlinked markdown knowledge base that grows in value over time.

Core Features & Use Cases

  • Wiki compilation instead of per-query RAG: ingests sources once into a structured wiki so future questions reuse compiled knowledge rather than re-extracting from scratch.
  • Three-layer architecture (raw → pages → schema): keeps immutable raw sources separate from agent-owned entity/concept/comparison/query pages, governed by a single SCHEMA.md.
  • Compounding consistency via orientation and logging: reads SCHEMA.md, index.md, and recent log.md at session start to avoid duplicates, maintain conventions, and track updates.

Quick Start

Start the LLM-wiki by telling the agent: "Create a new wiki at my configured path, define its SCHEMA for AI research notes, and ingest these sources into the wiki while updating index and log."

Frequently Asked Questions about llm-wiki

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

FAQPage Schema
How do I build a persistent knowledge base from markdown sources instead of using per-query RAG?▼

You can build a compounding markdown wiki by ingesting sources once into a structured, interlinked knowledge base with a three-layer architecture, which eliminates the need for per-query RAG extraction on repeated research.

How do I ingest URLs and PDFs into an interlinked markdown wiki?▼

Ingest URLs and PDFs by instructing the agent to compile them into entity and concept pages with required frontmatter, while keeping the original raw sources immutable and tracking provenance through citations.

What is the best way to track provenance and citations in a markdown knowledge base?▼

Track provenance by maintaining a three-layer architecture that separates immutable raw source files from agent-owned pages, using citations within the markdown to link compiled knowledge directly back to the original raw source files.

How do I lint and audit the health of an interlinked markdown wiki?▼

Audit wiki health by linting for required frontmatter, validating cross-references between entity pages, verifying index and log updates, and checking provenance citations against the stored raw source files.

Can I use a markdown wiki to avoid duplicate research and scattered notes?▼

Yes, a markdown wiki solves scattered notes by applying orientation and logging at session start, reading the schema and index to avoid duplicate entries, maintain conventions, and track updates as the knowledge base compounds.

Related Skills

Explore More Education & Research Skills→
SebasArenSebasAren
2

obsidian-wiki-ingest

Classify raw materials and generate wiki pages from PDFs, notes, and URLs.

Community
Advanced
unclehowellunclehowell

llm-wiki

Create and maintain a persistent interlinked markdown knowledge base.

Community
Advanced
koasis89koasis89

wiki

Manage persistent markdown knowledge bases with keyword search and lifecycle tracking.

Community
Intermediate
DarkArty07DarkArty07
3

llm-wiki

Build persistent interlinked markdown knowledge bases with provenance tracking and wikilinks.

Community
Advanced