skills/nick-vels/velsvisual/visual/Gen Agent Trust Hub

visual

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill guides the agent to install the velsvisual utility from the NPM registry and execute setup scripts using npx.
  • [COMMAND_EXECUTION]: The skill relies on shell command execution for its core functionality, using the velsvisual CLI to manage models and run media generation tasks.
  • [DATA_EXFILTRATION]: The velsvisual upload command enables the transmission of local files to the external kie.ai service for processing.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because user-provided prompts are passed directly into shell commands without explicit sanitization. 1. Ingestion points: User prompts in SKILL.md. 2. Boundary markers: No delimiters or ignore instructions are present for user input. 3. Capability inventory: Execution of shell commands via the velsvisual CLI. 4. Sanitization: No sanitization or escaping of shell metacharacters is instructed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 04:26 PM
Security Audit — agent-trust-hub — visual