commercial

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the genmedia CLI tool for key operations including model searches (genmedia models), asset uploads (genmedia upload), and generation execution (genmedia run). These commands are standard for the skill's media production workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing user-supplied data that could lead to indirect prompt injection. 1. Ingestion points: User-defined product characteristics, goals, and brand constraints in SKILL.md. 2. Boundary markers: No delimiters or instructions to ignore embedded commands are included in the prompt templates. 3. Capability inventory: The agent is instructed to use the genmedia tool for network operations, file uploads, and disk writes (SKILL.md). 4. Sanitization: No sanitization logic is provided to handle potentially malicious shell characters in user inputs before they are passed to the --prompt argument.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:40 PM
Security Audit — agent-trust-hub — commercial