llm-wiki

Compile source documents into an interlinked Obsidian-style wiki with append-only logs.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It solves the problem of turning scattered notes, research, and source documents into a consistently linked, continuously updated knowledge base without relying on one-off answers or fragile retrieval.

Core Features & Use Cases

  • Compile-first knowledge accumulation: incrementally ingests one source at a time and updates wiki pages so knowledge grows and stays interlinked.
  • Obsidian-ready wiki structure: organizes immutable inputs in raw/, LLM-owned markdown in wiki/, and an evolving schema in CLAUDE.md or AGENTS.md.
  • Entity and concept wikilinking: converts named entities and concepts into [[Wikilink]] references to maintain navigable connectivity.
  • Auditability and quality control: uses wiki/log.md for an append-only history and provides an on-demand lint flow to detect orphans, hallucinated claims, broken links, and contradictions.

Quick Start

Initialize your vault using this skill’s template, place your first source document into raw/, then say "ingest" to create the first set of wiki pages and update the index.

Frequently Asked Questions about llm-wiki

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

FAQPage Schema
How do I compile scattered research notes into a linked Obsidian wiki?

To compile scattered research notes into a linked Obsidian wiki, the skill incrementally ingests source documents into an immutable raw folder and generates interlinked markdown pages with wikilinks and frontmatter summaries.

What is the best way to maintain an Obsidian knowledge base without per-query retrieval?

Maintaining an Obsidian knowledge base without per-query retrieval involves using a compile-first pipeline that builds a persistent wiki state, deriving answers from interlinked entity pages instead of fragmented search queries.

How does wikilinking work for entities when building a personal knowledge management wiki?

Wikilinking for entities in a personal knowledge management wiki works by converting named entities and concepts extracted from source documents into navigable [[Wikilink]] references within the generated markdown files.

Can I audit my Obsidian vault for broken links and hallucinated claims?

Yes, you can audit your Obsidian vault for broken links and hallucinated claims using an on-demand lint flow that scans the compiled wiki state to detect orphans, contradictions, and broken references via an append-only log.

Does this research workflow require a specific folder structure for source documents?

Yes, this research workflow requires a strict folder structure separating immutable inputs in a raw directory from compiled markdown in a wiki directory, alongside a schema file for consistent page generation.

Why use a compile-first wiki approach instead of standard RAG for research notes?

A compile-first wiki approach replaces standard RAG for research notes by maintaining a continuously updated, interlinked knowledge state, avoiding the fragile retrieval and one-off answers associated with per-query lookups.