One-click install
npx skills add https://github.com/dbsdud/llm-wiki-skill --skill wiki-dbsdud
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wiki
Source: https://github.com/dbsdud/llm-wiki-skill/tree/main
Command: npx skills add https://github.com/dbsdud/llm-wiki-skill --skill wiki-dbsdud

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you reliably capture new knowledge and continuously curate it into an Evergreen, citation-backed personal LLM wiki instead of re-searching and re-synthesizing every time.

Core Features & Use Cases

  • Vault bootstrapping and schema: Initializes a consistent vault structure at ~/vaults with CLAUDE.md workflow and frontmatter conventions.
  • Ingestion with synthesis: Ingests URLs, files, and project knowledge (via graphify outputs) into immutable raw/ sources and generates curated wiki/sources summaries plus selective concept/entity pages.
  • Retrieval with citations and maintenance: Answers questions using wiki/index and related pages with provenance, and runs linting to find orphans, broken sources, frontmatter issues, and stale pages.

Quick Start

Run /wiki init to create your ~/vaults wiki, then ingest a URL by asking /wiki to ingest-url https://example.com/blog and summarize it into your vault.

Frequently Asked Questions about wiki

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

FAQPage Schema
How do I build a personal knowledge vault from scattered sources?

To build a personal knowledge vault, initialize a consistent schema at ~/vaults and ingest URLs, local files, and project outputs. The system maintains an immutable raw-to-curated workflow, generating summaries and selective concept pages with YAML frontmatter.

What is the best way to maintain an LLM wiki with citations?

The best way to maintain an LLM wiki with citations is through ongoing vault maintenance, which includes running linting to find orphans, broken sources, and frontmatter issues. New pages must be registered in wiki/index.md and activity logs appended in wiki/log.md.

How does knowledge management ingestion work with URLs and local files?

Knowledge management ingestion works by capturing URLs and local files into an immutable raw/ sources directory. The system then synthesizes this data into curated wiki/sources summaries and concept pages, enforcing a strict raw-to-curated workflow.

Can I answer questions using my personal knowledge base with provenance?

Yes, you can answer questions using your personal knowledge base with provenance. The system retrieves information from wiki/index and related pages, ensuring that answers are citation-backed and traceable to the original ingested sources.

Does graphify integration support ingesting project outputs into a wiki?

Yes, graphify integration supports ingesting project outputs directly into the wiki vault. The ingested outputs are processed through the standard immutable raw-to-curated workflow, generating curated summaries and entity pages with citations.

Why does vault maintenance require appending logs without modifying past entries?

Vault maintenance requires appending logs without modifying past entries to preserve an immutable activity history. This ensures consistent provenance tracking and prevents the corruption of previous synthesis states within the knowledge management system.