dream-video-prompt-generator
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of markdown documentation and instructional templates (
SKILL.md,references/*.md,assets/*.md). No executable scripts (.py, .js, .sh) or configuration files that trigger code execution are included. - [PROMPT_INJECTION]: Instructions focus on character consistency and quality control for prompt generation. There are no attempts to bypass safety filters, extract system prompts, or override agent behavior using malicious injection patterns.
- [DATA_EXFILTRATION]: The skill does not contain any network-related commands (e.g.,
curl,wget,fetch), hardcoded credentials, or access to sensitive file paths (e.g.,.env,.ssh). - [REMOTE_CODE_EXECUTION]: No package managers, remote script downloads, or dynamic code execution patterns were detected. The skill operates purely within the text-generation context.
- [COMMAND_EXECUTION]: No shell command execution or subprocess spawning patterns were found in the instructional text.
- [SAFE]: The structured workflow includes a 'Quality Inspector' role, which acts as a verification layer to ensure output consistency, representing a safe architectural pattern for agent-based tasks.
Audit Metadata