earth2studio-discover

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill follows best practices for a discovery-focused agent extension.
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches documentation from official NVIDIA domains (nvidia.github.io and github.com/NVIDIA). These are well-known technology service providers and the downloads are limited to documentation files for informational purposes.
  • [DATA_EXPOSURE]: The skill does not access sensitive local files or hardcode credentials. It requests task-specific information from the user only to provide filtered recommendations.
  • [COMMAND_EXECUTION]: There are no instructions for the agent to execute shell commands, install software, or run code. The skill's instructions explicitly exclude these actions from its scope.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a low-risk surface as it ingests external data from documentation URLs (Ingestion points: documented URLs in SKILL.md). However, it possesses no exploitable capabilities (Capability inventory: None) like file writing or command execution, and uses structured citation rules (Boundary markers: Step 4 and 6), which prevents the ingested content from influencing the system in a harmful way.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 04:34 PM
Security Audit — agent-trust-hub — earth2studio-discover