Create and maintain interlinked markdown knowledge bases with linting and Obsidian integration.
npx skills add https://github.com/graniet/kheish --skill llm-wiki-graniet
Please help me install this Agent Skill. Skill: llm-wiki Source: https://github.com/graniet/kheish/tree/main/skills/research/llm-wiki Command: npx skills add https://github.com/graniet/kheish --skill llm-wiki-graniet
Traditional RAG systems rediscover knowledge from scratch for every query, leading to redundant processing, missed cross-references, and inconsistent insights. Manually maintaining a structured, searchable knowledge base is time-consuming, and in-process agent knowledge is lost when the caller crashes or disconnects, resulting in repeated work and fragmented information.
Use the llm-wiki skill to ingest the attached arxiv paper on mixture of experts and create linked pages for all mentioned models and techniques.
High-intent search queries and answers about installing and using this skill.
To build a persistent markdown wiki that compounds research insights, ingest raw sources like URLs and PDFs into an immutable layer, then synthesize cross-linked entity and concept pages to retain knowledge over time without redundant processing.
Cross-link validation works by running automated linting tools that detect orphan pages, broken wikilinks, and stale content across your markdown knowledge base, ensuring structural consistency and enforcing tag taxonomy rules.
Yes, you can use Obsidian to view and query a generated markdown wiki because the system syncs seamlessly with Obsidian, supporting visual graph views, headless server setups, and Dataview queries for cross-referenced research content.
The best way to ingest arxiv papers and generate linked pages is to ingest the source documents into an immutable raw layer, then automatically synthesize and cross-link entity, concept, and comparison pages for all mentioned models and techniques.
You detect contradictory claims and taxonomy violations across a markdown wiki by utilizing built-in consistency auditing and linting tools that automatically scan for orphan pages, broken links, stale content, and conflicting statements.
Traditional RAG misses cross-references because it rediscoveries knowledge from scratch for every query, leading to redundant processing and inconsistent insights, whereas a structured wiki architecture compiles and cross-links content persistently.
lookatitudeIngest URLs, files, or pasted text into categorized wiki pages with provenance metadata.
tylerburleighDetect and repair structural issues in SDD JSON specification files.
markwang2658Ingest sources into a persistent interlinked markdown wiki with schema compliance.
SignmanalIngest sources into an interlinked markdown knowledge base with wikilinks and provenance markers.
