nuget
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to a remote Model Context Protocol (MCP) server at
https://unbrowse.ai/mcp/nuget.org. This server provides the tool definitions and handles the interaction with the target website (nuget.org) server-side. - [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes external content from nuget.org search results. As it processes third-party data, it is a surface for indirect prompt injection.
- Ingestion points: Search results and page content returned by the
nuget_org__get_packagesandnuget_org__get_searchtools. - Boundary markers: Absent; there are no explicit delimiters or instructions provided to the agent to distinguish between the site content and its own command context.
- Capability inventory: The skill has the ability to perform network operations through the MCP service and perform browsing actions like autofilling or clicking (SKILL.md).
- Sanitization: The skill returns readable text and links; no specific sanitization or filtering of the results is mentioned for security purposes.
Audit Metadata