skills/doany-ai/skills/relight/Gen Agent Trust Hub

relight

Pass

Audited by Gen Agent Trust Hub on May 29, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the installation of the @runcomfy/cli package via npm and the use of npx. These are standard methods for accessing the vendor-provided tool required for the skill's functionality.
  • [COMMAND_EXECUTION]: Execution is limited to the runcomfy CLI using the Bash tool. The skill explicitly addresses shell injection risks by utilizing JSON for input parameters, ensuring that untrusted prompt content is not interpreted as shell commands.
  • [PROMPT_INJECTION]: The skill is designed to handle external image URLs and user-defined prompts. This constitutes an indirect prompt injection surface. The documentation proactively provides guidance on mitigating these risks, such as limiting processing to user-provided assets and monitoring for unexpected behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
May 29, 2026, 11:27 AM
Security Audit — agent-trust-hub — relight