layer-reference-sets

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill mentions the installation of sibling skills using the command npx skills add layerai/skills. This refers to the official repository of the skill author (layerai) for extending agent capabilities.
  • [COMMAND_EXECUTION]: Instructions include the use of the npx skills add CLI command to manage the agent's installed skills and functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided datasets, including images and potentially associated metadata, to train custom models.
  • Ingestion points: User-supplied image folders and files provided during the curation and training process.
  • Boundary markers: No explicit delimiters are specified for the processing of user-provided content metadata.
  • Capability inventory: The skill utilizes tools for model training (start_training) and image generation (execute_forge).
  • Sanitization: No specific sanitization or filtering logic is detailed within the prompt instructions; the skill relies on the platform's internal handling of dataset assets.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:27 AM
Security Audit — agent-trust-hub — layer-reference-sets