transformers-js

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing external data (text, images, and audio) through machine learning models.\n
  • Ingestion points: Untrusted data is ingested through the pipeline API parameters across SKILL.md and references/EXAMPLES.md.\n
  • Boundary markers: No explicit delimiters are used to isolate data from model instructions.\n
  • Capability inventory: The skill is capable of model inference and writing to a local filesystem cache.\n
  • Sanitization: Standard machine learning guardrails are assumed; no manual sanitization is implemented.\n- [EXTERNAL_DOWNLOADS]: The skill retrieves resources from well-known and trusted external services.\n
  • Fetches machine learning models from Hugging Face's official repository.\n
  • Loads the Transformers.js library via the JSDelivr CDN in browser-based examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:52 AM
Security Audit — agent-trust-hub — transformers-js