llm-wiki

Compile raw sources into a persistent structured wiki of markdown pages.

85|15|Updated Apr 15, 2026
One-click install
npx skills add https://github.com/praneybehl/llm-wiki-plugin --skill llm-wiki-praneybehl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: llm-wiki
Source: https://github.com/praneybehl/llm-wiki-plugin/tree/main/skills/llm-wiki
Command: npx skills add https://github.com/praneybehl/llm-wiki-plugin --skill llm-wiki-praneybehl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill solves the context-bottleneck problem of accumulating and querying large document collections. Traditional RAG systems re-derive knowledge from raw chunks on every query, but this skill compiles sources once into a persistent, structured wiki of markdown pages, allowing knowledge to compound over time without hitting context limits or losing prior synthesis.

Core Features & Use Cases

  • Ingest & Compile: Process papers, articles, transcripts, and notes into structured wiki pages with cross-references, entity tracking, and source citations.
  • Query & Synthesize: Answer questions against the accumulated knowledge base and file answers back as synthesis pages so future queries benefit.
  • Lint & Maintain: Run structural and semantic health checks to catch contradictions, orphan pages, broken links, and stale claims before they compound.
  • Scale to Thousands: Sharded indexes, atomic page sizing, YAML frontmatter, and BM25 search keep the wiki usable from dozens to thousands of pages.

Quick Start

Use the llm-wiki skill to ingest the attached document into your wiki and ask what it says about the topic.

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 knowledge base from research papers without hitting LLM context limits?

Build a knowledge base from research papers without hitting context limits by compiling raw sources once into a persistent, structured wiki of markdown pages. This allows knowledge to compound across documents without re-reading raw sources on every query.

How does a compounding wiki differ from traditional RAG for accumulating research notes?

A compounding wiki differs from traditional RAG by synthesizing raw sources into persistent markdown pages with cross-references and citations upfront. RAG re-derives knowledge from raw chunks on every query, whereas the wiki files answers back as synthesis pages for future use.

Can I scale a markdown knowledge base to thousands of pages and still search it efficiently?

You can scale a markdown knowledge base to thousands of pages by using sharded indexes, atomic page sizing, YAML frontmatter conventions, and bundled BM25 search scripts to maintain query performance and structural health.

How do I ingest articles and transcripts into a structured wiki with cross-references?

Ingest articles and transcripts into a structured wiki by processing raw sources into markdown pages with YAML frontmatter, cross-references, entity tracking, and source citations to ensure knowledge is synthesized and connected.

What is the best way to maintain a large knowledge base and catch contradictions or broken links?

Maintain a large knowledge base and catch contradictions by running structural and semantic health checks. These linting checks identify orphan pages, broken links, stale claims, and contradictions before they compound across the wiki.

Do I need YAML frontmatter and pyyaml to compile documents into a searchable wiki?

You need YAML frontmatter and the pyyaml dependency to compile documents into a searchable wiki, as these conventions maintain the structured metadata required for sharded indexes and BM25 search scripts to function at scale.