Public Opinion Scanner
PassAudited by ClawScan on May 17, 2026.
Overview
This instruction-only skill coherently searches public web sources and creates a local opinion report, with no code, install step, credentials, or hidden behavior shown.
This skill appears safe to install as an instruction-only public web research workflow. Be aware that it sends target names and related terms to search/fetch providers, and verify the generated sentiment labels and sources before using or sharing the report.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may send the target company/person name and related search terms to external search or fetch services and retrieve public pages.
The skill directs the agent to perform multiple external searches and fetch pages. This is central to the opinion-scanning purpose and is bounded, but users should know it will actively query/fetch public web sources.
对目标执行至少 6 组搜索... 每条搜索获取 **5-10 条结果**... 对每条有效结果，用 `web_fetch` 或 `tavily_extract` 获取详细内容
Use it only for targets and queries you are comfortable sending to web/search providers, and review the collected sources before relying on the report.
The generated report may contain reputational information about people or organizations and will remain in the workspace until removed.
The skill persists a report containing summarized public information and sentiment labels about a company or person. This is expected output, but it creates a stored artifact that could later be shared or reused.
报告默认保存到：`/root/.openclaw/workspace/reports/opinion/{target-slug}/{YYYY-MM-DD_HHmm}.md`Check the saved report for accuracy and sensitivity before sharing it, and delete it when it is no longer needed.