Multi Search Engine 2

Multi search engine integration with 17 engines (8 CN + 9 Global). Supports advanced search operators, time filters, site search, privacy engines, and WolframAlpha knowledge queries. No API keys required.

This skill appears to do what it says (construct and fetch public search URLs) and requires no credentials or install. Still, consider the following before installing: 1) Provenance: the publisher and file metadata disagree (owner ID and version mismatches) and there's no homepage or repo — prefer skills with clear authorship. 2) Misuse potential: the documentation includes examples (e.g., intext:password, filetype:txt, cache:) that make it easy to craft OSINT queries to find leaked credentials or other sensitive info; think about whether you want an agent with autonomous web_fetch ability performing such queries. 3) If you proceed, limit autonomous invocation or monitor web_fetch outputs, avoid supplying sensitive or private keywords to the skill, and prefer installing only from verified sources or repositories with traceable authorship. If you want higher assurance, ask the publisher for a source repo or signed release or choose a similar skill from a known maintainer.

Type: OpenClaw Skill Name: multi-search-engine-2 Version: 1.0.0 The skill is designed for multi-engine web searching, which is a benign purpose. However, the `references/international-search.md` file contains explicit examples of advanced search operators that demonstrate how to find sensitive information, specifically `intext:password filetype:txt`. While this is a legitimate OSINT technique and the skill only instructs the agent to construct the search URL via `web_fetch` (not to exfiltrate or exploit the results), providing such an example constitutes a risky capability without clear malicious intent, thus classifying it as suspicious rather than benign.