Menu
Search the web for facts and current information. Brave Search (requires API key), Wikipedia and DuckDuckGo (free). Tools: http.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
Search, subscribe, and listen to podcasts. Download RSS feeds, manage episodes, and play audio with position tracking. Tools: http, file_storage, media_queue, play_audio.
Manage local lists (shopping, to-do, etc.) stored as files. Add, remove, check and read items. Tools: file_storage.
Read, create, update and delete Google Calendar events. Requires Google OAuth. Tools: http.
Search contacts by name and retrieve phone numbers. Use before calling or sending SMS. Tools: query.
Read, search and send emails via Gmail API. Requires Google OAuth. Tools: gmail_send, http.
Start navigation, get directions and route info via Google Maps app. Tools: intent, http (optional, requires Maps API key for traffic/route details).
| name | web-research |
| description | Search the web for facts and current information. Brave Search (requires API key), Wikipedia and DuckDuckGo (free). Tools: http. |
Search the web, look up facts, read web pages, and synthesize information from multiple sources.
| Tool | Purpose |
|---|---|
http | GET/POST requests to search APIs and web pages |
check_credential | Verify if Brave Search API key is configured |
file_storage | Save research results for later reference |
scheduler | Schedule recurring research tasks |
Brave Search provides comprehensive web results with snippets. Free tier: 2000 queries/month.
Always check the API key first:
{
"tool": "check_credential",
"args": { "credential_id": "brave_search_api_key" }
}
If configured: false → fall back to Wikipedia/DuckDuckGo (see below). Do NOT attempt the Brave API call.
{
"method": "GET",
"url": "https://api.search.brave.com/res/v1/web/search?q={QUERY}&count=5",
"auth_provider": "brave_search_api_key",
"auth_header": "X-Subscription-Token"
}
Response fields:
web.results[] – array of search results
.title – page title.url – page URL.description – snippet/summary (HTML stripped).age – how old the result is (e.g. "2 hours ago")query.original – the original search query{
"method": "GET",
"url": "https://api.search.brave.com/res/v1/news/search?q={QUERY}&count=5",
"auth_provider": "brave_search_api_key",
"auth_header": "X-Subscription-Token"
}
Response fields:
results[] – array of news articles
.title – article title.url – article URL.description – summary snippet.age – publication time (e.g. "3 hours ago").meta_url.hostname – source domain (e.g. "bbc.com")Perfect for factual lookups, definitions, biographies, historical events, and encyclopedic queries.
{
"method": "GET",
"url": "https://en.wikipedia.org/api/rest_v1/page/summary/{TOPIC}",
"headers": [{ "key": "User-Agent", "value": "SannaBot/1.0" }]
}
Response fields:
title – article titleextract – plain text summary (first paragraph)description – short descriptioncontent_urls.desktop.page – full article URLLanguage support: Replace en with the target language code:
https://de.wikipedia.org/api/rest_v1/page/summary/{TOPIC}https://fr.wikipedia.org/api/rest_v1/page/summary/{TOPIC}https://es.wikipedia.org/api/rest_v1/page/summary/{TOPIC}https://it.wikipedia.org/api/rest_v1/page/summary/{TOPIC}{
"method": "GET",
"url": "https://en.wikipedia.org/w/api.php?action=query&list=search&srsearch={QUERY}&utf8=&format=json&srlimit=5",
"headers": [{ "key": "User-Agent", "value": "SannaBot/1.0" }]
}
Response fields:
query.search[] – array of results
.title – article title (use this with the summary endpoint).snippet – text excerpt (contains <span> tags – strip HTML)Workflow: Search → pick best matching title → fetch summary via REST API.
Provides quick factual answers, definitions, and related topics. Best for simple "what is X?" queries.
{
"method": "GET",
"url": "https://api.duckduckgo.com/?q={QUERY}&format=json&no_html=1&skip_disambig=1"
}
Response fields:
AbstractText – short answer/summary (often from Wikipedia)AbstractSource – source nameAbstractURL – source URLAnswer – direct answer (e.g. for calculations, conversions)RelatedTopics[] – list of related topics
.Text – topic description.FirstURL – link to more infoNote: If both AbstractText and Answer are empty, DuckDuckGo has no instant answer → fall back to Brave Search or Wikipedia.
Use the http tool with response_format: "text" to fetch and read a specific URL:
{
"method": "GET",
"url": "{PAGE_URL}",
"response_format": "text",
"headers": [{ "key": "User-Agent", "value": "SannaBot/1.0" }]
}
Important:
| User intent | Best source | Fallback |
|---|---|---|
| "Search for …" / "Google …" / "Look up …" | Brave Search | Wikipedia search |
| "What is …?" / "Who is …?" / "Define …" | Wikipedia summary | DuckDuckGo Instant |
| "Latest news about …" | Brave News search | Headlines skill (RSS) |
| "Tell me about [person/place/concept]" | Wikipedia summary | Brave Search |
| "How does [X] work?" | Wikipedia + Brave Search | DuckDuckGo |
| "Compare X and Y" | Wikipedia (both) + Brave | – |
| "Read this article: [URL]" | Direct http fetch | – |
| Quick fact / number / date | DuckDuckGo Instant | Wikipedia |
check_credential("brave_search_api_key")http → GET https://api.search.brave.com/res/v1/web/search?q={TOPIC}&count=5 with auth_provider: "brave_search_api_key", auth_header: "X-Subscription-Token"web.results[]: extract title, description, URL for each resulthttp → GET Wikipedia search API for {TOPIC}http → GET https://en.wikipedia.org/api/rest_v1/page/summary/{X} (use de.wikipedia.org for German users, etc.)extract fieldcheck_credential("brave_search_api_key")http → GET https://api.search.brave.com/res/v1/news/search?q={TOPIC}&count=5 with authhttp calls)http → GET {URL} with response_format: "text"+ or %20: artificial+intelligence& → %26, # → %23, ? → %3FZürich, Münchenhttp tool truncates responses at 5000 characters – this is enough for search results and Wikipedia summaries