image-prompt-director
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided creative ideas to generate detailed image prompts, creating an attack surface for indirect prompt injection. * Ingestion points: User-provided one-line ideas (SKILL.md, references/full-prompt.md). * Boundary markers: None defined for the input. * Capability inventory: The skill generates text and lacks dangerous capabilities like file system access, network exfiltration, or code execution (SKILL.md). * Sanitization: No explicit sanitization or filtering of user input is documented.
- [EXTERNAL_DOWNLOADS]: The README.md provides instructions to download the skill and its dependencies from the author's GitHub repository using npx, git, and python scripts. These represent standard installation steps from the vendor's infrastructure.
- [COMMAND_EXECUTION]: The installation guide in README.md lists shell commands for users to execute to set up the skill environment. These commands are for manual user setup and are not triggered by the skill itself during execution.
Audit Metadata