Ingest raw sources into an indexed, append-only markdown wiki.
npx skills add https://github.com/HybridAIOne/hybridclaw --skill llm-wiki-hybridaione
Please help me install this Agent Skill. Skill: llm-wiki Source: https://github.com/HybridAIOne/hybridclaw/tree/main/skills/llm-wiki Command: npx skills add https://github.com/HybridAIOne/hybridclaw --skill llm-wiki-hybridaione
This Skill solves the problem of maintaining a durable, self-updating knowledge base by integrating new raw sources into an organized wiki over time instead of regenerating answers from scratch each turn.
wiki/sources/, and update related entities, concepts, and analyses while preserving cross-references.index.md as the catalog of what exists and append structured activity entries to log.md to track create, ingest, query artifacts, lint passes, and archives.AGENTS.md as the workflow and frontmatter contract, including link style, taxonomy, thresholds, and conflict handling, to prevent drift and duplicate pages.Ask the assistant to ingest the attached document into your LLM wiki at the specified wiki root, updating summaries, relevant entities, and the index with an append-only log entry.
High-intent search queries and answers about installing and using this skill.
A markdown wiki ingests raw sources incrementally by summarizing new inputs, updating related entities and concepts, and logging all changes to an append-only knowledge layer to maintain coherence over time.
Maintain an append-only knowledge base by using a schema-driven workflow that reads current state before writing, updating cross-references and an index while logging all file changes to prevent duplicate pages and drift.
Entity extraction during ingest works by reading the existing wiki schema, summarizing the new source, and synthesizing extracted entities and concepts into maintained wiki pages while preserving cross-references.
Yes, you can prevent duplicate pages by honoring a three-layer layout separating raw immutable inputs, maintained wiki pages, and AGENTS governance files to enforce taxonomy, link style, and conflict handling thresholds.
Yes, you need to perform orientation reads before updating your research knowledge base to check the current index and schema state, preventing data drift and ensuring new ingests merge cleanly with existing wiki pages.
You should not use an incremental ingest workflow for source summaries if your project requires regenerating answers from scratch each turn instead of maintaining a persistent, cross-referenced, and audit-friendly markdown wiki structure.
pax-kDistill raw documents into an Obsidian wiki with frontmatter metadata and wikilinks.
ccomkhjIngest raw sources into immutable raw/ files and compile structured wiki articles.
inbrainfunIngest video, audio, PDF, book, screenshot, and GitHub repo sources into a brain.
samuelmukotiIngest sources into a persistent Markdown wiki with YAML frontmatter and wikilinks.