Answer questions against an LLM wiki knowledge base. Use this skill whenever the user asks a question that requires searching and synthesizing information from the wiki. The LLM searches for relevant pages, reads them, and synthesizes an answer with citations. Answers can take different forms: markdown page, comparison table, slide deck (Marp), chart, or canvas. Important: good answers can be filed back into the wiki as new pages. Perfect for: research questions, comparing concepts, generating summaries, creating analyses, or any question that builds on accumulated wiki knowledge. Always use this when user mentions query, ask, question, compare, analyze, or wants information from the wiki.
This skill answers questions against an existing LLM-powered wiki knowledge base following the LLM Wiki pattern.
Query is the operation where the LLM searches for relevant pages in the wiki, reads them, and synthesizes an answer with citations. The key insight: good answers should be filed back into the wiki as new pages so explorations compound.
Start by reading index.md to find relevant pages. The index provides:
Based on the question, identify which pages are relevant:
Read the identified pages to gather information. Note:
Combine information from multiple sources:
Choose appropriate format based on question type:
If the answer is valuable and reusable, create a new wiki page:
Example: `markdown
Present the answer with:
`markdown
[2-3 sentence answer]
`markdown
| Aspect | [A] | [B] |
|---|---|---|
| [Aspect 1] | [Detail] | [Detail] |
| [Aspect 2] | [Detail] | [Detail] |
| ` |