Skip to main content

google-search

Search the web using Google via AceDataCloud API. Use when searching for web pages, images, news, maps, local places, or videos. Supports localization, time filtering, and pagination. Returns structured results with titles, snippets, URLs, and rich data.

インストールへ移動

ソース情報

リポジトリ
AceDataCloud/Skills
ソースの最終更新活動
2026年8月9日 14:38
検出された SKILL.md の言語
英語
スター
17
フォーク
1

インストール方法

デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。

ソースファイルを確認

インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。

SKILL.md を表示中

SKILL.md
ソースの指示 · 読み取り専用プレビュー
name
google-search
description
Search the web using Google via AceDataCloud API. Use when searching for web pages, images, news, maps, local places, or videos. Supports localization, time filtering, and pagination. Returns structured results with titles, snippets, URLs, and rich data.
license
Apache-2.0
metadata
{"author":"acedatacloud","version":"1.0"}
compatibility
Requires ACEDATACLOUD_API_TOKEN in .env file (see _shared/authentication.md). Optionally pair with mcp-serp for tool-use.
# Google Search (SERP) Search the web through AceDataCloud's Google SERP API. > **Setup:** See [authentication](../_shared/authentication.md) for token setup. ## Quick Start ```bash curl -X POST https://api.acedata.cloud/serp/google \ -H "Authorization: Bearer $ACEDATACLOUD_API_TOKEN" \ -H "Content-Type: application/json" \ -d '{"query": "latest AI news", "type": "search"}' ``` ## Search Types | Type | Description | Returns | |------|-------------|---------| | `search` | Web search (default) | Organic results, knowledge graph, rich snippets | | `images` | Image search | Image URLs, titles, sources | | `news` | News articles | Headlines, sources, publish dates | | `maps` | Map results | Locations, coordinates | | `places` | Local businesses/places | Name, address, rating, reviews | | `videos` | Video results | Video URLs, thumbnails, duration | ## Parameters ```json POST /serp/google { "query": "your search query", "type": "search", "country": "us", "language": "en", "range": "qdr:w", "number": 10, "page": 1 } ``` | Parameter | Type | Description | |-----------|------|-------------| | `query` | string | Search query (required; must contain non-whitespace text; maximum 2048 characters) | | `type` | string | One of: search, images, news, maps, places, videos | | `country` | string | Non-empty country code (e.g., "us", "uk", "cn", "jp"; maximum 32 characters) | | `language` | string | Non-empty language code (e.g., "en", "zh", "ja"; maximum 32 characters) | | `range` | string | Time filter (see below) | | `number` | int | Number of results per page | | `page` | int | Page number for pagination | | `image_size` | string | **Images only.** Filter by size for high-res sources: `large` / `medium` / `icon`, or a megapixel minimum `2mp`…`70mp` (e.g. `4mp` = larger than 4 megapixels). Use `large` (or a `*mp` value) whenever the image will be shown large / full-screen / zoomed. | ## Time Range Options | Value | Period | |-------|--------| | `qdr:h` | Past hour | | `qdr:d` | Past 24 hours | | `qdr:w` | Past week | | `qdr:m` | Past month | | `qdr:y` | Past year | | `h` | Past hour | | `d` | Past 24 hours | | `w` | Past week | | `m` | Past month | | `y` | Past year | ## Response Structure Web search returns structured data including: - `organic_results`: Main search results with title, link, snippet - `knowledge_graph`: Entity information panel (when available) - `related_searches`: Related query suggestions ## Gotchas - Default search type is `"search"` (web). Always specify `type` for non-web searches - Country and language codes affect result localization significantly - `number` controls results per page, not total results — use `page` for pagination - Time range (`range`) only applies to web search and news, not images or places - **Image resolution (important for video / full-screen use):** results include `image_url` (full-size), `thumbnail_url`, and `image_width`/`image_height`. Pass **`image_size: "large"`** (or a megapixel minimum like `"4mp"`) to get sharp sources, and pick the result with the largest `image_width`×`image_height`. Always download `image_url` — **never** use `thumbnail_url` as a final asset (it is tiny and blurry). - Places search works best with location-specific queries (e.g., "restaurants near Times Square") > **MCP:** `pip install mcp-serp` | Hosted: `https://serp.mcp.acedata.cloud/mcp` | See [all MCP servers](../_shared/mcp-servers.md)
GitHubで見る