编写、调试与调优 Playwright 测试,涵盖定位器策略、追踪诊断与 CI 友好的超时配置。
作者 ivangdavila1.3k 次安装130 星标
浏览器
Google News API - Headlines by Keyword, Topic, Publication
试用Search Google News for headlines by keyword, topic, or publication as structured JSON — headline, source, date, and link. Use for current events, monitoring, and news research. v2 engine, 1 credit per request.
它能做什么
**✅ Works Powerfully & Reliably On All Major AI Assistants**
技能文档
Google News Automation Scraper Skill
✨ Platform Compatibility
✅ Works Powerfully & Reliably On All Major AI Assistants
| Platform | Status | How to Install |
|---|---|---|
| OpenCode | ✅ Fully Supported | Copy skill folder to ~/.opencode/skills/ |
| Claude Code | ✅ Fully Supported | Native skill support |
| Cursor | ✅ Fully Supported | Copy to ~/.cursor/skills/ |
| OpenClaw | ✅ Fully Supported | Compatible |
Why Choose BrowserAct Skills?
- 🚀 Stable & crash-free execution
- ⚡ Fast response times
- 🔧 No configuration headaches
- 📦 Plug & play installation
- 💬 Professional support
📖 Introduction
This skill provides a one-stop news collection service using BrowserAct's Google News API template. It allows the agent to retrieve structured news data with a single command.
🔑 API Key Guidance
Before running, check the BROWSERACT_API_KEY environment variable. If not set, do not proceed with script execution; instead, request the API key from the user.
Required Message to User:
"Since you haven't configured the BrowserAct API Key, please go to the BrowserAct Console to get your Key and provide it to me in this chat."
🛠️ Input Parameters
Flexibly configure these parameters based on user requirements:
-
Search_Keywords
- Type:
string - Description: Keywords to search on Google News (e.g., company names, industry terms).
- Example:
AI Startup,Tesla,SpaceX
- Type:
-
Publish_date
- Type:
string - Description: Time range filter for articles.
- Options:
any time: No restrictionpast hours: Breaking newspast 24 hours: Daily monitoring (Recommended)past week: Short-term trendspast year: Long-term research
- Default:
past week
- Type:
-
Datelimit
- Type:
number - Description: Maximum news items to extract.
- Default:
30 - Suggestion: Use 10-30 for monitoring, higher for research.
- Type:
🚀 Execution (Recommended)
Execute the following script to get results:
# Call Example
python .cursor/skills/google-news-api/scripts/google_news_api.py "Keywords" "TimeRange" Count
📊 Data Output
Successful execution returns structured data:
headline: News titlesource: Publishernews_link: URLpublished_time: Timestampauthor: Author name (if available)
⚠️ Error Handling & Retry Mechanism
-
Check Output:
- If output contains
"Invalid authorization", the API Key is invalid. Do not retry. Guide the user to provide a correct key. - For other failures (e.g.,
Error:or empty results), automatically retry once.
- If output contains
-
Retry Limit:
- Maximum one automatic retry. If it still fails, stop and report the error to the user.
相关技能
Persistent browser and Electron interaction through `js_repl` for fast iterative UI debugging.
作者 OpenAI28.2k 星标
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser l
作者 Anthropic182.7k 星标
Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrappe
作者 OpenAI28.2k 星标
Henk Nie 的更多技能
浏览全部技能通过托管 OAuth 接入 GitHub REST API，覆盖仓库、Issue、PR、Commit、分支与用户等接口。
作者 scavio-ai1 次安装
Pull a Facebook page's profile, posts, reels and photos, one post with its comments, a single reel/video with downloadable URLs, a public group and its posts, an event, and hashtag posts. 11 endpoints, 1 credit each, structured JSON.
作者 scavio-ai3 次安装
Search the Meta Ad Library by keyword or Facebook Page id and walk full cursor pagination through every ad, with creative, run dates, platforms and political spend. 3 endpoints, 1 credit per page.
作者 scavio-ai3 次安装
Search YouTube and retrieve videos, shorts, comments, transcripts, streams, channel data, and community posts with their comments as structured JSON. 17 endpoints across video and channel surfaces.
作者 scavio-ai1 次安装
Search live or SOLD eBay listings, read a listing in full, and look up a seller's profile card as structured JSON. 3 endpoints, 1 credit each. Sold-listing search is the price-research feature eBay itself hides.
作者 scavio-ai1 次安装
Search Google Hotels for a destination and dates, then fetch per-property vendor pricing and full details — as structured JSON. Price, rating, class, and amenity filters. v2 engine, 1 credit per request.
作者 scavio-ai1 次安装