effectcraft-kinetic-type

Warn

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documents a download location for a portable ZIP archive containing an executable (effectcraft-cli.exe) from a third-party repository (github.com/storytold).
  • [EXTERNAL_DOWNLOADS]: The skill references an external API endpoint (api.skillhub.cn) for downloading motion graphics assets.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-controlled text to generate animations through the layer.newText command, creating a surface for indirect instructions.
  • Ingestion points: User-supplied text strings are interpolated into the text parameter of animation tool calls in SKILL.md and references/craft-methods.md.
  • Boundary markers: The instructions do not define delimiters or provide warnings to the agent to ignore potential instructions embedded within the text content.
  • Capability inventory: The skill possesses the ability to execute engine commands via command_run, write project files using save_project, and inspect output with inspect across multiple scripts.
  • Sanitization: No sanitization, escaping, or validation logic is defined to process the input text before it enters the tool execution pipeline.
  • [DYNAMIC_EXECUTION]: Employs dynamic command construction for the effectcraft-cli tool and command_run interface, where parameters are mapped from templates at runtime.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 23, 2026, 09:37 AM