primevideo
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill configures a remote Model Context Protocol (MCP) server at
https://unbrowse.ai/mcp/primevideo.com. This connects the agent to a third-party service managed by the skill author (Unbrowse.ai) to process requests and fetch data from Prime Video. - [INDIRECT_PROMPT_INJECTION]: The skill's primary purpose is to ingest and display content from external websites (
primevideo.comandamazonvideo.com). - Ingestion points: Tools
primevideo_com__get_searchandprimevideo_com__read_page(defined inSKILL.md) fetch title, text, and links from external web pages. - Boundary markers: The instructions do not specify any delimiters or safety warnings for the agent to ignore potential instructions embedded in the fetched web content.
- Capability inventory: The skill includes tools for searching, reading pages, and performing browser-like actions via the Unbrowse platform (
unbrowse.run,unbrowse.browse.act). - Sanitization: There is no evidence of sanitization or filtering of the external content before it is presented to the agent.
Audit Metadata