te-path-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data from the ThousandEyes API (such as hop-by-hop DNS names, BGP prefix metadata, and routing details) and provides 'AI-powered explanation' of these results, creating an attack surface where malicious network-level data could influence the agent's interpretation.
- Ingestion points: Network diagnostic data including path visualizations, BGP route details, and event metrics fetched from the ThousandEyes API (SKILL.md).
- Boundary markers: None specified for the interpolation of external network data into the agent's explanatory context.
- Capability inventory: Tools for path visualization, BGP analysis, instant test execution, and anomaly detection (SKILL.md).
- Sanitization: No sanitization of external network content (e.g., DNS names or BGP attributes) is described before processing.
- [EXTERNAL_DOWNLOADS]: The skill references and fetches tool definitions from a remote MCP server located at
https://api.thousandeyes.com/mcp. This is a connection to an official, well-known service associated with Cisco ThousandEyes for the purpose of network troubleshooting.
Audit Metadata