seedance-meme-comedy
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user-provided descriptions (setup, twist, subject) and interpolate them into a fixed prompt structure for video generation.
- Ingestion points: Inputs are collected during the initial workflow (Step 1) and via the provided copy-paste templates in
SKILL.md. - Boundary markers: The skill uses bracketed placeholders (e.g.,
[Setup: ...]) for user data but lacks explicit negative instructions or sanitization to prevent the user-provided text from influencing the agent's behavior. - Capability inventory: Analysis of
SKILL.mdandreferences/cases.mdshows no tools, subprocess calls, network requests, or file-system write operations. - Sanitization: No escaping or filtering is applied to external content before interpolation.
Audit Metadata