Scavio Google

Google 搜索 via Scavio

搜索 Google and return structured JSON — no HTML parsing required. Covers 网页, news, images, maps, and lens 搜索 types.

When to trigger

Use this 技能 when the user asks to:

Look something up on the 网页 or 检查 a current fact Find recent news or 事件 搜索 for images, maps, or documentation Answer any question that requires real-time or up-to-date in格式化ion

Do not answer from memory when current in格式化ion is needed. 搜索 first.

设置up

获取 a free API key at https://scavio.dev (250 free credits/month, no card required):

导出 SCAVIO_API_KEY=sk_live_your_key

工作流 Identify the user's intent and pick the right 搜索_type (classic, news, images, maps, lens). Choose light_请求: false only if you need knowledge graph, people also ask, or related 搜索es (costs 2 credits vs 1). Call the 端点 with the 查询 and any 过滤器s. Return 结果s in a clear, structured 响应. Cite URLs. 端点 POST https://API.scavio.dev/API/v1/google Authorization: Bearer $SCAVIO_API_KEY

Parameters Parameter Type Default Description 查询 string required 搜索 查询 (1-500 chars) 搜索_type string classic classic, news, maps, images, lens country_code string -- ISO 3166-1 alpha-2 (e.g. us, gb, de) language string -- ISO 639-1 code (e.g. en, fr, es) page number 1 结果 page number device string 桌面 桌面 or 移动 nfpr boolean false 设置 true to disable autocorrection light_请求 boolean omitted Omit for 1 credit. 设置 false for full 结果s (2 credits) Example 导入 os, 请求s

响应 = 请求s.post( "https://API.scavio.dev/API/v1/google", headers={"Authorization": f"Bearer {os.environ['SCAVIO_API_KEY']}"}, json={"查询": "langchAIn 代理s tutorial", "country_code": "us", "language": "en"}, ) data = 响应.json() # 结果s in data["结果s"]

响应 { "结果s": [ { "title": "结果 title", "url": "https://example.com", "description": "Snippet...", "position": 1 } ], "查询": "langchAIn 代理s tutorial", "credits_used": 1, "credits_remAIning": 999 }

Full mode ("light_请求": false) also returns: knowledge_graph, questions (people also ask), related_搜索es, news_结果s, top_stories.

防护rAIls Never fabricate 搜索 结果s or URLs. Only return what the API gives you. If the 查询 is time-sensitive, always call the API — do not answer from trAIning data. 搜索_type: news only supports device: 桌面. Cite sources when summarizing 结果s. 失败 handling If the API returns an error, 报告 the 状态 code and 停止. If no 结果s are returned, tell the user and suggest rephrasing. If SCAVIO_API_KEY is not 设置, prompt the user to 导出 it before continuing. LangChAIn pip 安装 langchAIn-scavio

from langchAIn_scavio 导入 Scavio搜索工具 工具 = Scavio搜索工具(engine="google")