fiftyone-zoo-remote-model

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The README.md file contains an installation command curl -sL skil.sh | sh -s -- voxel51/fiftyone-skills. This is the standard mechanism for installing agent skills on this platform and points to the vendor's own repository, presenting no elevated risk in this context.
  • [SAFE]: The skill facilitates processing external data (images and prompts) via model inference. The architectural guidance provided in VLM-PATTERNS.md and LABEL-TYPES.md includes mitigation strategies for handling untrusted model outputs, such as using multi-tier fallback parsers and strict coordinate normalization, which effectively manages the attack surface for indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 11:56 AM
Security Audit — agent-trust-hub — fiftyone-zoo-remote-model