ndr-de
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the external website ndr.de, creating a surface for indirect prompt injection attacks where malicious instructions could be embedded in the site's content.
- Ingestion points: Tools such as
ndr_de__get_search,ndr_de__get_suche, andndr_de__read_page(defined inSKILL.md) fetch content from ndr.de. - Boundary markers: The instructions lack explicit delimiters or safety prompts advising the agent to ignore instructions found within the retrieved web data.
- Capability inventory: The skill includes capabilities like
unbrowse.runandunbrowse.browse.act(referenced inSKILL.md) which could be abused if the agent follows instructions injected from an external page. - Sanitization: There is no documentation of content filtering or sanitization being applied to the data fetched from ndr.de.
- [EXTERNAL_DOWNLOADS]: The skill connects to a remote Model Context Protocol (MCP) server provided by the author to execute its functions.
- Evidence: The
server.jsonfile specifies a remote endpoint athttps://unbrowse.ai/mcp/ndr.deusing thestreamable-httptransport.
Audit Metadata