Help us improve
Share bugs, ideas, or general feedback.
From wikipedia-knowledge-agent
This skill should be used when the user asks to "look up a Wikipedia article", "search Wikipedia for", "find the Wikipedia page on", "check Wikipedia about", or needs a quick fact or single article from the local Wikipedia data. For quick lookups and single-article reads — not for cross-domain exploration.
npx claudepluginhub timbrinded/wikipedia-knowledge-agentHow this skill is triggered — by the user, by Claude, or both
Slash command
/wikipedia-knowledge-agent:wiki-lookupThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Quick lookup skill for the local Wikipedia data lake (~7M articles stored as plain text).
Search and query the wiki — get cited answers, with automatic research when knowledge is missing. Three depth modes. Use when: 'check wiki', 'what do we know about', 'look up', 'find in wiki', 'search wiki', 'search for', 'find information about', 'do we have notes on', 'wiki context', 'existing knowledge', 'research', 'look into', 'find out about', 'investigate', 'what is', 'how does X work'.
Implements 6-step deterministic ladder to check Karpathy wiki coverage for any user question: orient via schema/index, count signal matches, inline-read candidates or spawn Explore subagent or web search.
Search the LLM Wiki using qmd (hybrid BM25/vector search). Finds relevant wiki pages, reads them, and synthesizes an answer with citations. Use when asking questions against the knowledge base.
Share bugs, ideas, or general feedback.
Quick lookup skill for the local Wikipedia data lake (~7M articles stored as plain text).
rg -i "<query>" data/index/titles.txt | head -20
rg -i "<query>" data/index/categories.txt | head -20
rg -i -w "<query>" data/index/titles.txt | head -20
Use -w when a short query returns too many irrelevant hits — it matches whole words only.
rg -i -e "<term1>" -e "<term2>" data/index/titles.txt | head -20
Search for multiple terms at once. Great for synonyms or when the article title might use different phrasing.
Find the path first, then read it:
rg -m1 "^<slug>\t" data/index/paths.txt | cut -f3
Then use the Read tool on the returned path.
head -50 data/articles/<prefix>/<slug>.txt
Scan the opening to decide if the article is worth a full read.
rg -l -i "<query>" data/articles/ | head -20
Returns file paths of matching articles. Narrow with title/category search first — full-text is powerful but slower.
rg -i -m2 -C1 "<query>" data/articles/<prefix>/<slug>.txt
-m2 limits to first 2 matches, -C1 shows 1 line of context around each. Useful for checking relevance before a full read.
rg -i -c "<query>" data/articles/ | sort -t: -k2 -nr | head -10
Articles with more mentions are likely more relevant.
rg -l -i "<term1>" data/articles/ | xargs rg -l -i "<term2>" | head -10
rg -F -i "<literal>" data/index/titles.txt
Use -F when your query contains regex-special characters (dots, parens, brackets).
Task: "What does Wikipedia say about circuit breaker patterns?"
rg -i "circuit.breaker" data/index/titles.txtrg -m1 "^circuit-breaker\t" data/index/paths.txt | cut -f3Task: "Find information about ant colony optimization"
rg -i "ant" data/index/titles.txt | head -20 — too many results (elephant, antenna, etc.)rg -i -w "ant" data/index/titles.txt | head -20 — better but still noisyrg -i "ant colony" data/index/titles.txt — found itrg -l -i "ant colony optimization" data/articles/ | head -5head -50 data/articles/an/ant-colony-optimization.txthead -50 saves time on irrelevant articles-w if too many results, broaden with -e synonyms if too fewIf the task requires cross-domain connections, non-obvious structural analogues, or synthesising insights across multiple articles from different fields — spawn the wiki-explorer agent instead. That agent is purpose-built for lateral knowledge exploration.
Examples of when to escalate: