seedance-prompter

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted user input (textual intent and asset metadata) to construct prompts for a downstream model, creating a vulnerability surface for indirect prompt injection.\n
  • Ingestion points: User-provided creative intent and the filenames of uploaded assets referenced in SKILL.md.\n
  • Boundary markers: The instructions lack explicit delimiters or specific 'ignore embedded instructions' warnings for the downstream model when handling these inputs.\n
  • Capability inventory: The skill's output is limited to a structured JSON object containing text prompts and numeric parameters; it has no file system, network, or shell execution capabilities.\n
  • Sanitization: No validation or sanitization of input strings (such as asset filenames) is mentioned in the logic.\n- [NO_CODE]: The skill consists entirely of markdown configuration and reference documents; no executable scripts, binaries, or package dependencies were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 12:15 PM