multi-search-engine-refined
v1.0.0Multi search engine integration with 4 engines: Google, DuckDuckGo, Brave Search, and WolframAlpha. Supports advanced search operators, time filters, and kno...
⭐ 0· 57·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
The skill's stated purpose (multi-engine search via Google, DuckDuckGo, Brave, WolframAlpha) matches the SKILL.md, config.json and references: all examples generate search URLs and call web_fetch. Nothing in the package requests unrelated permissions or credentials. However, registry-level metadata (skill name/version/owner) does not fully match the embedded files (_meta.json, metadata.json, config.json use 'multi-search-engine' and version 2.0.1) while the top-level registry block shows version 1.0.0 and a different owner ID — this mismatch reduces trust and should be checked with the publisher.
Instruction Scope
SKILL.md instructs only to construct search URLs for the four allowed domains and call web_fetch on them; it does not ask the agent to read local files, environment variables, or send data to other endpoints. It explicitly prefers DuckDuckGo's HTML endpoint to avoid browser redirects. The instructions are scoped to search-related tasks.
Install Mechanism
This is an instruction-only skill with no install spec and no code to write to disk, which is low-risk. All provided files are documentation/config JSON; nothing will be downloaded or executed during install.
Credentials
No environment variables, API keys, or credential/config paths are requested. The lack of secrets aligns with the claimed 'no API keys required' functionality.
Persistence & Privilege
Skill is not marked always:true and does not request elevated persistence. It is user-invocable and can be called autonomously by the agent (platform default), which is expected for a skill of this type.
Assessment
This skill is an instruction-only set of docs and templates for building search URLs for Google, DuckDuckGo, Brave, and WolframAlpha and does not request credentials or install software — that makes it reasonably low-risk. Before installing: 1) verify the publisher/owner and version (registry metadata and embedded _meta.json/metadata.json differ here), 2) confirm your agent's web_fetch implementation does not automatically follow redirects or send fetched content to third parties, and 3) avoid passing sensitive credentials or private data into search queries (search pages can leak parts of queries in logs). If you need stronger assurance, ask the publisher to reconcile the metadata mismatches or run the skill in a sandboxed agent first.Like a lobster shell, security has layers — review code before you run it.
latestvk975fptpcn1kzkgckp9xewb9d584a42x
License
MIT-0
Free to use, modify, and redistribute. No attribution required.