LLM Wiki Query

SkillDocs & knowledge

Answer questions from the compiled Wiki. Use whenever the user asks a knowledge question that this vault might cover.

Available today. Use it from your connected AI after setup.

No other account needed.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the LLM Wiki Query skill

What this skill tells your AI

The instructions your AI receives, as published by gavishap/omnia-vault in .claude/skills/llm-wiki-query/SKILL.md and read by ahel’s review.

Answer from compiled knowledge first; open Raw sources last.

Search the catalog:

python scripts/wiki_tool.py search-catalog --query "<user topic>"

Open the most relevant compiled Wiki notes (usually 1–3).
Answer from the compiled notes. Only open files under Raw/Sources/ when:
- the compiled note is insufficient, or
- the user asks for source-level verification.
When the answer depends on source material, cite both the compiled note and the Raw source path.

Do not scan Raw/Sources/ wholesale — that defeats the purpose of the Wiki.
If the catalog has no match, say so, then optionally check Schema/source-manifest.jsonl for unprocessed sources that might cover it.
If you find a gap worth filling, suggest an ingest rather than answering from memory.

Advanced