TSMAI Web Search
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @user_ccf72f3e/tsmai-search.
About this skill
Problem
When an agent workflow needs current external facts, model knowledge can be stale and search API outputs can be inconsistent. TSMAI Web Search wraps the Tencent Cloud SearchPro API as a callable tool for conversations, RAG pipelines, or automated tasks that need up-to-date web information.
How It Works
The skill centers on one search call: it reads Tencent Cloud API credentials from environment variables, runs the search script to send the request, and supports optional parameters for query control. The response is organized as Markdown, which makes it easier to pass into a prompt, notes, or downstream parsing. For integration work, this bundles web lookup, authentication handling, and result formatting in one layer.
Boundaries
This capability depends on valid Tencent Cloud API keys and network access to the service. Results are affected by SearchPro indexing, availability, and quota. It provides search summaries and structured Markdown output, not full-page scraping, deep reading, or factual verification. During debugging, verify that credentials work, request parameters match the API contract, and the output format meets downstream display needs.
Use Cases
- When an agent conversation needs current pricing, call it to run one search and read the Markdown summary.
- To add external facts to an assistant, format SearchPro results into Markdown blocks ready for the prompt.
- When checking reported online status changes, search related summaries and inspect the returned structure for announcements.
- When building a lightweight search helper, use its script as the backend lookup endpoint and handle the Tencent Cloud key.
Best For
- Engineers maintaining agent conversations who need to add current web facts to model context.
- Developers integrating Tencent Cloud APIs who need search results rendered as Markdown.
- Builders creating assistants who need external summaries merged into response context.
- Engineers debugging external search scripts who need to verify credentials and output structure.
Related Skills
Local workflow memory with matching and SOP updates.
An OpenClaw live streaming executor that initializes TRTC streaming, starts a real-time dashboard, generates viewer URLs, and continuously reports live events.
Breaks down physical supply chains for super-trends to identify second- and third-layer bottlenecks, runs valuation and reverse checks, and maintains trackable reports.
A token-saving compression mode for Chinese LLMs with lite, full, ultra, and classical tiers, preserving code and technical terms while handling edge cases.