modelscope

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli package via npm. This is a vendor-owned tool provided by the skill author for interacting with their platform.
  • [COMMAND_EXECUTION]: The skill uses the membrane CLI to perform various operations, including authentication, connection management, and running machine learning workflows. These commands are executed in the agent's shell environment.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes and interprets data from external ModelScope resources.
  • Ingestion points: Data retrieved from actions and model/dataset metadata via the membrane CLI (SKILL.md).
  • Boundary markers: Absent. No specific delimiters or instructions to ignore embedded content are provided for the processed data.
  • Capability inventory: Local shell command execution via the membrane CLI and external API interaction (SKILL.md).
  • Sanitization: Absent. The skill does not explicitly describe sanitization or validation of the data fetched from ModelScope.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 05:38 AM
Security Audit — agent-trust-hub — modelscope