Multi Search Engine.Tmp
v1.0.0Multi search engine integration with 17 engines (8 CN + 9 Global). Supports advanced search operators, time filters, site search, privacy engines, and Wolfra...
⭐ 0· 56·0 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The declared purpose (multi-search integration) matches the provided URLs, examples, and config.json. Minor inconsistencies: SKILL.md references references/advanced-search.md which is not present in the manifest, some metadata ownerId values differ across files, and the registry name is 'Multi Search Engine.Tmp' while the internal name is 'multi-search-engine'. One search URL choice (Bing INT using cn.bing.com with enseach=1) looks odd but not inherently malicious.
Instruction Scope
The runtime instructions are limited to building search URLs and calling web_fetch against public search engine endpoints. There are no instructions to read local files, access environment variables, or transmit data to private endpoints. The examples include advanced search operators (e.g., intext:password) which could be used for OSINT or sensitive-data discovery — this is capability-appropriate but potentially privacy-sensitive depending on user behavior.
Install Mechanism
No install spec and no bundled executable code — this is an instruction-only skill so it does not write or execute downloaded code on the host. That gives it a low install-surface risk.
Credentials
The skill requests no environment variables, credentials, or config paths. That is proportionate for a search-URL/HTTP-fetching documentation skill.
Persistence & Privilege
always is false and there are no indications the skill modifies other skills or agent-wide settings. It therefore does not request elevated persistence or privileges.
Assessment
This skill is an instruction-only collection of search URLs and examples and does not request secrets or perform local file access. Before enabling: (1) confirm you trust the unknown publisher (homepage is missing and owner IDs vary across metadata files); (2) be aware web_fetch will make outbound HTTP requests — ensure that is acceptable for your environment and privacy needs; (3) avoid running queries that intentionally probe for sensitive data (e.g., intext:password, filetype:txt searching for credentials); and (4) verify any slightly odd URLs (e.g., Bing INT pointing at cn.bing.com) or missing reference files if you rely on completeness of documentation.Like a lobster shell, security has layers — review code before you run it.
latestvk976jz2f8g5t9aymz3a5fh0s0s83p7r9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.