Multi Search Engine

**Summary:** This version updates cookie handling and search execution policies for increased privacy and more ethical crawling. - Switched to initializing an empty in-memory cookie store; no cookies are pre-loaded from files. - Cookies are acquired dynamically only when a search request is denied and cleared immediately after the session. - Only relevant search engines are queried, based on language and engine availability, rather than always searching all engines in a category. - Introduced stricter rate limiting (1-2 second delays, batch sizes of 3-4) to control request flow. - Strengthened privacy statements: cookies are held strictly in memory and never written to disk or config files. - Improved documentation of search logic, cookie management, and crawling ethics.

No changelog provided.

No changelog provided.

OpenClaw analysis

The skill's requirements and runtime instructions are internally consistent with a multi-engine search aggregator: it is instruction-only, requests no credentials or installs, and its behavior (fetching search result pages, obtaining session cookies in-memory to recover from 403/429, applying rate limits) aligns with the described purpose.

Confidence: high

VirusTotal

Type: OpenClaw Skill Name: multi-search-engine Version: 2.1.3 The skill bundle is a legitimate search aggregator supporting 16 domestic and international search engines. The instructions in SKILL.md and the documentation in the references directory focus on constructing standard search queries, implementing rate limiting, and managing session cookies strictly in memory to handle rate-limiting (403/429 errors). No evidence of data exfiltration, malicious execution, or harmful prompt injection was found; the logic is entirely consistent with the stated purpose of providing a multi-engine search interface.