gaijin
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to a remote streamable-http MCP server at
https://unbrowse.ai/mcp/gaijin.netto provide its core tool functionality. It also references an OpenAPI specification athttps://unbrowse.ai/api/v1/sites/gaijin.net/openapi.json. These resources are hosted on the vendor's own infrastructure. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: Tools such as
gaijin_net__read_pageandstore_gaijin_net__get_search_phpfetch title, text, and links from external gaijin.net pages into the agent's context. - Boundary markers: The skill instructs the agent to only report results when a
statusofsucceededis returned, which serves as a basic operational guideline. - Capability inventory: The skill includes capabilities for network-proxied interactions through tools like
unbrowse.browse.actandunbrowse.run. - Sanitization: There are no explicit instructions for the agent to sanitize or escape the content retrieved from gaijin.net before processing it.
Audit Metadata