據 AlphaMoat 統計,截至 2026-09,xAI Search 累計下載 3,866 次、獲 1 個星標,在全部 5,078 個 AI 技能中排名第 2,693 位,在「知識管理」分類(442 個)中排名第 246 位。
| 平台 | 累計安裝 | 累計下載 | 星標 | 評論 |
|---|---|---|---|---|
| ClawHub ↗ | 128 | 3682 | 1 | - |
| SkillHub ↗ | 240 | 3866 | 1 | — |
各平台的計數口徑不同,統計時間不一致。
該技能透過 xAI 的 Grok API 即時搜尋 X/Twitter 和網頁,利用 grok-3-fast 等模型的 web_search 與 x_search 智慧體工具,支援 curl 直接呼叫或 scripts/xai-search.py 輔助指令碼,可按賬號、日期範圍、圖片/影片理解等過濾 X 搜尋,按域名過濾 Web 搜尋;突發新聞與輿情用 X 搜尋,事實與研究類查詢用 Web 搜尋或兩者並用。
資料釋出於 2026-10-27
以下是根據技能文件生成的示例,不代表實際執行結果。
2026-09 起統計方法調整:同名技能按作者分別統計,並覆蓋全部技能。圖中虛線為調整前的月份,與之後的資料不直接比較,也不計算跨越調整的增長。
| 月份 | 累計下載 | 環比 | 星標 |
|---|---|---|---|
| 2026-09 | 3866 | — | 1 |
| 2026-08 | 3629 | +1.8% | 1 |
| 2026-07 | 3565 | +5.3% | 1 |
| 2026-06 | 3387 | +2.8% | 1 |
| 2026-05 | 3295 | — | 1 |
| 2026-03 | 2972 | — | 1 |
以下是作者釋出的原始檔案,按純文字顯示,本站未做修改或稽核;其中的連結不可點選,命令請核實後再執行。抓取於 2026-10-23。
---
name: xai-search
description: Search X/Twitter and the web in real-time using xAI's Grok API with agentic search tools.
metadata: {"clawdbot":{"emoji":"🔍"}}
---
# xAI Search (Grok API)
Use xAI's agentic search to query X/Twitter and the web in real-time. This leverages Grok's `web_search` and `x_search` tools.
**Docs:** https://docs.x.ai/docs/
## Requirements
- `XAI_API_KEY` environment variable
- Python 3 + xai-sdk: `pip install xai-sdk`
## Quick Usage (curl)
### Web Search
```bash
curl -s https://api.x.ai/v1/chat/completions \
-H "Authorization: Bearer $XAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-3-fast",
"messages": [{"role": "user", "content": "YOUR QUERY HERE"}],
"tools": [{"type": "function", "function": {"name": "web_search"}}]
}' | jq -r '.choices[0].message.content'
```
### X/Twitter Search
```bash
curl -s https://api.x.ai/v1/chat/completions \
-H "Authorization: Bearer $XAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-3-fast",
"messages": [{"role": "user", "content": "YOUR QUERY HERE"}],
"tools": [{"type": "function", "function": {"name": "x_search"}}]
}' | jq -r '.choices[0].message.content'
```
### Combined (Web + X)
```bash
curl -s https://api.x.ai/v1/chat/completions \
-H "Authorization: Bearer $XAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-3-fast",
"messages": [{"role": "user", "content": "YOUR QUERY HERE"}],
"tools": [
{"type": "function", "function": {"name": "web_search"}},
{"type": "function", "function": {"name": "x_search"}}
]
}' | jq -r '.choices[0].message.content'
```
## Helper Script
For convenience, use the `xai-search.py` script in the `scripts/` folder:
```bash
# Web search (adjust path to your skill location)
python ~/.clawdbot/skills/xai-search/scripts/xai-search.py web "latest news about AI"
# X/Twitter search
python ~/.clawdbot/skills/xai-search/scripts/xai-search.py x "what are people saying about Clawdbot"
# Both
python ~/.clawdbot/skills/xai-search/scripts/xai-search.py both "current events today"
```
## Models
- `grok-3-fast` — fast, good for quick searches
- `grok-4-1-fast` — reasoning model, better for complex queries
## X Search Filters
You can filter X searches by:
- `allowed_x_handles` / `excluded_x_handles` — limit to specific accounts
- `from_date` / `to_date` — date range (ISO8601 format)
- `enable_image_understanding` — analyze images in posts
- `enable_video_understanding` — analyze videos in posts
## Web Search Filters
- `allowed_domains` / `excluded_domains` — limit to specific sites
- `enable_image_understanding` — analyze images on pages
## Tips
- For breaking news: use X search
- For factual/research queries: use web search or both
- For sentiment/opinions: use X search
- The model will make multiple search calls if needed (agentic)
把長文字、文件與網頁內容壓縮成要點和關鍵詞的摘要工具。
透過 IMA OpenAPI 管理筆記與知識庫內容。
規定如何為網頁請求選擇合適工具的必讀指引。
呼叫百度 AI 搜尋介面做網頁檢索的技能。
用 uvx 直接呼叫 markitdown,把各類檔案轉成 Markdown。
面向 OpenClaw 的預設網頁搜尋,即時檢索多個引擎。
用 DuckDuckGo 介面做網頁、新聞、圖片與影片檢索的技能。
檢索並發現可用 OpenClaw 技能的技能。
資料月份:2026-09 · 下載/星標/安裝數聚合自公開技能註冊庫(ClawHub、SkillHub),按月更新。口徑說明見資料方法。