comic-ip-onboarder

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security threats detected. The skill focuses on character consistency and image generation workflows within an isolated environment.
  • [COMMAND_EXECUTION]: The skill does not use shell commands, subprocess calls, or system-level operations. It operates entirely within the standard agent interaction model and restricts the agent to using specific image generation tools while explicitly forbidding web-based rendering hacks (HTML/SVG).
  • [DATA_EXFILTRATION]: No network access or data transmission patterns were found. The skill manages output locally using a project-specific directory structure (ip-profiles/) for character profile creation.
  • [PROMPT_INJECTION]: The skill uses detailed instructional templates to guide the agent, but it does not contain patterns intended to bypass safety filters, extract system prompts, or override agent constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 03:14 AM
Security Audit — agent-trust-hub — comic-ip-onboarder