novelai-workflows
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to process user-provided strings and pass them directly to generation tools.
- Ingestion points: Prompt parameters in
generate_image,image_to_image, anddirector_toolwithinSKILL.md. - Boundary markers: Absent. No delimiters or instructions are provided to the agent to isolate user input from system instructions.
- Capability inventory: The skill involves executing shell commands via
novelai-image-mcp, writing files to the local disk, and making network requests to NovelAI API endpoints. - Sanitization: Absent. There is no guidance on escaping or validating user inputs.
- [COMMAND_EXECUTION]: The skill describes how to chain tasks using shell commands, including variable interpolation (
$IMG,$DIR). If an agent uses these instructions to execute shell commands with untrusted input, it could lead to command injection vulnerabilities.
Audit Metadata