← 返回 Skills 市场
4894
总下载
7
收藏
23
当前安装
3
版本数
在 OpenClaw 中安装
/install serper
功能描述
Google search via Serper API with full page content extraction. Fast API lookup + concurrent page scraping (3s timeout). One well-crafted query returns rich results — avoid multiple calls. Two modes, explicit locale control. API key via .env.
安全使用建议
Install only if you are comfortable using a Serper API key and having search terms sent to Serper while result URLs are fetched from external sites. Avoid sensitive secrets, private internal URLs, or confidential investigations as queries, and treat extracted webpage text as untrusted research material.
功能分析
Type: OpenClaw Skill
Name: serper
Version: 3.0.2
The OpenClaw skill bundle for Serper is benign. The `scripts/search.py` script legitimately uses the Serper API for web search and `trafilatura` for content extraction, requiring network access to `google.serper.dev` and arbitrary web pages. It retrieves the `SERPER_API_KEY` from environment variables or a `.env` file, which is standard practice for API-based tools. The `SKILL.md` instructions guide the AI agent on efficient and responsible use of the skill, including defensive instructions to prevent redundant web fetching, rather than attempting malicious prompt injection. No evidence of data exfiltration, malicious execution, persistence, obfuscation, or supply chain attacks was found.
能力评估
Purpose & Capability
The artifacts coherently describe Google search through Serper plus full-page content extraction, and the Python script implements that behavior without unrelated capabilities.
Instruction Scope
Runtime instructions are scoped to when and how to search, locale selection, and avoiding redundant fetching; there is no hidden role change, prompt override, or unrelated workflow.
Install Mechanism
Installation is manual and user-scoped, but it asks users to install the unpinned third-party Python package trafilatura.
Credentials
The skill requires a Serper API key, sends search queries to google.serper.dev, and fetches returned URLs; this is proportionate for a search skill but has privacy implications for sensitive queries.
Persistence & Privilege
The code reads environment variables or a local .env for the API key and shows no background persistence, privilege escalation, broad local file access, destructive actions, or credential exfiltration.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install serper - 安装完成后，直接呼叫该 Skill 的名称或使用
/serper触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v3.0.2
- No code or documentation changes detected in this release.
- Version number update only; functionality and documentation remain unchanged.
v3.0.1
- Added LICENSE and README.md files for legal clarity and documentation.
- Updated documentation to emphasize the new fast API lookup and concurrent page scraping model (3-second timeout per page).
- Output format changed to streamed JSON array, with results emitted one at a time as pages finish.
- Stronger guidance to use only one well-crafted query per invocation for maximum efficiency—multiple sequential queries are discouraged.
- Clarified differences between modes, strict locale requirements, and best practices.
v3.0.0
Serper 3.0.0 — Major update with full page content extraction and explicit locale support
- Adds full-text page content extraction using trafilatura; not just snippets but entire article text from each result.
- Introduces two precise search modes: `default` for general/all-time queries and `current` for news/recent info.
- Requires explicit locale flags (`--gl`, `--hl`) for all non-English or location-specific searches.
- New CLI structure for flexible search modes and locale control.
- Output is clean per-line JSON: first with summary, then each full page/article as retrieved.
- Detailed usage guidance and strict do-not-use for overlapping web-fetch tools.
元数据
常见问题
serper 是什么？
Google search via Serper API with full page content extraction. Fast API lookup + concurrent page scraping (3s timeout). One well-crafted query returns rich results — avoid multiple calls. Two modes, explicit locale control. API key via .env. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 4894 次。
如何安装 serper？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install serper」即可一键安装，无需额外配置。
serper 是免费的吗？
是的，serper 完全免费（开源免费），可自由下载、安装和使用。
serper 支持哪些平台？
serper 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 serper？
由 nesdeq（@nesdeq）开发并维护，当前版本 v3.0.2。
推荐 Skills