yahoo-co-jp
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external websites (
news.yahoo.co.jp,auctions.yahoo.co.jp), creating an attack surface where malicious instructions embedded in web content (such as news titles, comments, or item descriptions) could influence the agent's behavior. - Ingestion points: The tools
news_yahoo_co_jp__get_search,auctions_yahoo_co_jp__get_search, andnews_yahoo_co_jp__read_pagedefined inSKILL.mdingest content from third-party sites. - Boundary markers: No specific delimiters or instructions to ignore embedded content were found in the skill definitions.
- Capability inventory: The skill utilizes network operations via an external MCP bridge to the vendor's API and includes capabilities like
unbrowse.runandunbrowse.browse.openfor broader site interaction. - Sanitization: There is no mention of explicit sanitization or filtering of the retrieved web content to prevent the execution of instructions contained within the data.
- [EXTERNAL_DOWNLOADS]: The skill connects to an external Model Context Protocol (MCP) server at
https://unbrowse.ai/mcp/yahoo.co.jp. This is a vendor-owned resource used to provide the skill's core functionality.
Audit Metadata