← 返回 Skills 市场
underbench2-gif

Web Research Assistant

作者 underbench2-gif · GitHub ↗ · v1.0.0
cross-platform ✓ 安全检测通过
1223
总下载
2
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install ub2-web-research-assistant
功能描述
Conducts structured web research by querying multiple sources, evaluating credibility, synthesizing data, and providing a detailed summary with citations.
使用说明 (SKILL.md)

Web Research Assistant

A skill that enables Claw to perform structured web research on any topic, synthesize findings from multiple sources, and produce a well-organized summary with citations.

What This Skill Does

This skill guides Claw through a systematic web research workflow:

  1. Query Formulation — Break down a broad research question into specific, targeted search queries
  2. Multi-Source Gathering — Search across multiple sources to collect diverse perspectives and data points
  3. Source Evaluation — Assess credibility and relevance of each source
  4. Synthesis — Combine findings into a coherent, structured summary
  5. Citation — Provide proper attribution with URLs for all referenced material

How to Use

Ask Claw to research any topic:

  • "Research the current state of quantum computing and summarize key breakthroughs from the past year"
  • "Find and compare the pricing models of the top 5 cloud providers"
  • "Investigate the environmental impact of electric vehicles vs. traditional cars"

Output Format

The skill produces a structured research report containing:

  • Executive Summary — A 2-3 sentence overview of key findings
  • Detailed Findings — Organized by subtopic with supporting evidence
  • Source List — All URLs and publication dates referenced
  • Confidence Assessment — How well-supported each finding is across sources

Best Practices

  • Be specific in your research question for more focused results
  • Specify a time range if you need recent information (e.g., "in the last 6 months")
  • Mention if you need a particular perspective (technical, business, consumer, etc.)
  • Request a specific output format if the default doesn't suit your needs
安全使用建议
This skill is instruction-only and internally consistent, so it poses low installation risk. Before you rely on outputs: (1) confirm your agent has safe, audited web-browsing/search capability (the skill assumes the agent can access the web); (2) avoid pasting secrets or private data into queries — the skill will synthesize and cite anything you provide; (3) verify citations and URLs the model returns (LLMs can hallucinate plausible-sounding sources or dates); (4) if you need authoritative coverage, specify allowed domains or time ranges in your prompt; and (5) run a few harmless test queries to confirm the quality and provenance of results before using it for sensitive decisions.
功能分析
Type: OpenClaw Skill Name: ub2-web-research-assistant Version: 1.0.0 The skill bundle contains only metadata (_meta.json) and a markdown file (SKILL.md). The SKILL.md file describes a 'Web Research Assistant' and provides instructions for the AI agent to perform structured web research, synthesize findings, and cite sources. There is no executable code, no evidence of prompt injection attempts, data exfiltration, unauthorized actions, or any other malicious or suspicious behavior. All content is consistent with the stated purpose.
能力评估
Purpose & Capability
Name and description match the actual requirements: it's an instruction-only research workflow that asks the agent to formulate queries, search sources, evaluate credibility, synthesize, and cite. There are no unrelated env vars, binaries, or config paths requested.
Instruction Scope
SKILL.md provides high-level, non-invasive instructions for conducting web research and source evaluation. It does not instruct the agent to read local files, environment variables, or transmit data to unexpected endpoints. Because the guidance is broad ("search across multiple sources"), actual behavior depends on the agent/platform browsing/search capabilities and any runtime safety policies; the instructions lack constraints (allowed/forbidden domains, privacy handling) which is a functional/usability note rather than an incoherence.
Install Mechanism
No install spec and no code files — lowest-risk arrangement. Nothing is downloaded or written to disk by the skill itself.
Credentials
The skill requires no environment variables, credentials, or config paths. Requested access is minimal and proportional to its stated purpose.
Persistence & Privilege
always:false and no indications the skill modifies agent/system settings or other skills. Autonomous invocation is allowed by default but is not combined here with any broad credentials or persistence, so no elevated privilege concerns.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install ub2-web-research-assistant
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /ub2-web-research-assistant 触发
  4. 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
v1.0.0
v1.0.0 - Initial release of Web Research Assistant skill. Enables structured web research with multi-source gathering, source evaluation, synthesis, and citation generation. Includes executive summary, detailed findings, and confidence assessment output formats.
元数据
Slug ub2-web-research-assistant
版本 1.0.0
许可证
累计安装 1
当前安装数 1
历史版本数 1
常见问题

Web Research Assistant 是什么?

Conducts structured web research by querying multiple sources, evaluating credibility, synthesizing data, and providing a detailed summary with citations. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1223 次。

如何安装 Web Research Assistant?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install ub2-web-research-assistant」即可一键安装,无需额外配置。

Web Research Assistant 是免费的吗?

是的,Web Research Assistant 完全免费(开源免费),可自由下载、安装和使用。

Web Research Assistant 支持哪些平台?

Web Research Assistant 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 Web Research Assistant?

由 underbench2-gif(@underbench2-gif)开发并维护,当前版本 v1.0.0。

💬 留言讨论