据 AlphaMoat 统计,截至 2026-09,Brave Web Search 累计下载 3,803 次、获 0 个星标,在全部 5,078 个 AI 技能中排名第 2,772 位,在「知识管理」分类(442 个)中排名第 256 位。
| 平台 | 累计安装 | 累计下载 | 星标 | 评论 |
|---|---|---|---|---|
| ClawHub ↗ | 126 | 3,803 | - | - |
| SkillHub ↗ | 122 | 3,760 | - | — |
各平台的计数口径不同,统计时间不一致。
它提供排序后的搜索结果(含 URL 和摘要片段),也可返回 AI 生成的简要答案。查询会发送到外部 API;避免在查询中输入敏感信息,并核实原始来源、时效和摘要准确性。 需要 Node.js 和相应 Brave API key;查询应作为独立参数传递,不拼接 shell 字符串。
资料发布于 2026-10-29
以下是根据技能文档生成的示例,不代表实际运行结果。
2026-09 起统计方法调整:同名技能按作者分别统计,并覆盖全部技能。图中虚线为调整前的月份,与之后的数据不直接比较,也不计算跨越调整的增长。
| 月份 | 累计下载 | 环比 | 星标 |
|---|---|---|---|
| 2026-09 | 3,803 | — | - |
| 2026-08 | 3,669 | +4.5% | - |
| 2026-07 | 3,511 | +9.5% | - |
| 2026-06 | 3,207 | +5.4% | - |
| 2026-05 | 3,042 | — | - |
| 2026-03 | 2,167 | — | - |
以下是作者发布的原始文件,按纯文本显示,本站未做修改或审核;其中的链接不可点击,命令请核实后再运行。抓取于 2026-10-23。
---
name: brave-web-search
description: Searches the web and returns ranked results or AI-generated summarized answers using the Brave Search API. Use for real-time web lookups and factual Q&A.
metadata:
clawdbot:
emoji: "🔍"
requires:
env: ["BRAVE_SEARCH_API_KEY", "BRAVE_ANSWERS_API_KEY"]
bins: ["node"]
primaryEnv: "BRAVE_SEARCH_API_KEY" category: "Search & Research"
---
# Brave Web Search
Searches the web and fetches AI-generated summarized answers using the Brave Search API. Exposes two commands: `brave-search` for ranked web results and `brave-answer` for concise AI summaries.
## Instructions
1. **Trigger**: Activate when the user wants to look something up on the web, check recent news, or get a factual answer to a question.
2. **Setup**: No installation step is required — this skill has zero external dependencies and runs on native Node.js.
3. **Command selection**:
- Use `brave-search` for general web searches where ranked results with URLs and snippets are useful.
- Use `brave-answer` for direct factual questions where a concise AI summary is more appropriate.
4. **Execution**: Invoke the script by passing the command name and parameters as **separate arguments**, never by interpolating user input into a shell command string. Use an argument array / `execFile`-style invocation so the shell never parses user-supplied values. Example (Node-style pseudo-code):
```javascript
execFile('node', ['index.js', 'brave-search', '--query', userQuery, '--count', '10'])
```
Do **not** construct the command as a single concatenated string such as `"node index.js brave-search --query " + userQuery`.
5. **Freshness**: For time-sensitive queries, pass `--freshness` followed by `pd` (past day), `pw` (past week), or `pm` (past month) as a separate argument to `brave-search`.
6. **Fallback**: If `brave-answer` returns `answer: null`, present the `fallback_results` to the user instead.
7. **Completion**: Present the results clearly, citing titles and URLs for web search results, or the summary text for answer results.
## Security & Privacy
- **Shell Injection Prevention**: User queries **must** be passed as discrete arguments (e.g. via `execFile` or an argv array), never interpolated into a shell command string. Concatenating user input into a shell string (e.g. `shell: true` with template literals) enables shell injection and is strictly forbidden.
- **Instruction Scope**: This skill only sends query strings to the Brave Search and Brave Summarizer APIs.
- **Environment**: It uses the `BRAVE_SEARCH_API_KEY` and `BRAVE_ANSWERS_API_KEY` provided by the OpenClaw environment.
- **Data Access**: It does not read local files or .env files. All configuration is handled by the agent.把长文本、文档与网页内容压缩成要点和关键词的摘要工具。
通过 IMA OpenAPI 管理笔记与知识库内容。
规定如何为网页请求选择合适工具的必读指引。
调用百度 AI 搜索接口做网页检索的技能。
用 uvx 直接调用 markitdown,把各类文件转成 Markdown。
面向 OpenClaw 的默认网页搜索,实时检索多个引擎。
用 DuckDuckGo 接口做网页、新闻、图片与视频检索的技能。
检索并发现可用 OpenClaw 技能的技能。
数据月份:2026-09 · 下载/星标/安装数聚合自公开技能注册库(ClawHub、SkillHub),按月更新。口径说明见数据方法。