Tencent Yuanbao Web Search
Paste the following prompt into your AI chat to install this skill:
Follow https://skillhub.cn/install/skillhub.md to install @user_f26e0ccb/tencent-yuanbao-standard-search.
About this skill
Problem to solve
When answering real-time or externally sourced questions about stocks, policy, news, or tutorials, model-only knowledge can be stale, incomplete, or hard to verify, and may lack source links. For prompts such as latest, 2024, or search for, the model needs fresh web evidence.
How the skill works
The skill builds a keyword search pipeline: it first checks whether TENCENTCLOUD_WSA_APIKEY is configured, then calls scripts/websearch.py with parameters such as query, site, cnt, freshness, and industry. freshness can be limited with d3, m2, or y1; industry can constrain results to gov, news, acad, or finance. Results are formatted as Markdown with linked titles, snippets, publish times, sites, relevance scores, and optional image information, making them easier for models to extract verifiable facts.
Boundaries and notes
It is suited to information lookup, trend tracking, knowledge retrieval, and answers that require source links. It is not for fully offline use and should not replace secondary verification of authoritative sources. Before use, the corresponding search service must be enabled and the environment variable configured correctly; if it reports unavailable, not enabled, or unauthorized, check billing status, service activation, and API key settings.
Use Cases
- When writing a tech blog, verify the latest API version changes by searching with keyword and time-range filters to get source links
- During investment research, query the latest financial news for a specific stock using freshness and finance parameters
- When drafting an industry report, collect recent authoritative media coverage of a policy document using industry and site filters
- When troubleshooting an outage, quickly find the latest community discussions of an error message by keyword, checking site and publish time
Best For
- Backend engineers integrating real-time web search into conversational AI products, needing external web evidence with source links
- Industry analysts regularly retrieving finance or policy news by time range and keyword, needing summarized material with site citations
- Technical documentation authors verifying API changes and best practices, needing high-quality snippets and URLs by site and keyword
- Academic researchers retrieving domain-specific English literature, using industry=acad and site parameters to scope authoritative sources
Related Skills
Provides Bazi, Liu Yao, Qimen, Meihua, and date-selection guidance from classical references, with plain-language explanations and actionable suggestions.
Extracts core themes and hierarchical relationships from documents, reports, web pages, and images into editable mind maps.
Parse WeChat favorites exports, classify them with keywords or LLM discovery, and export to Obsidian, Notion, or IMA.
Enter a video URL or local file and automatically download, transcribe, and analyze it into structured outputs such as summaries, quotes, chapters, subtitles, mind maps, and share cards.