anti-render
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes a process for ingesting and analyzing user-provided images to generate secondary prompts and images.
- Ingestion points: The workflow explicitly starts with '接收图像' (Receive Image) in
SKILL.md, taking external binary data as input. - Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard potential adversarial instructions embedded in image metadata or visual content.
- Capability inventory: The skill leverages
image-to-promptandprompt-to-imagetools to transform ingested data. - Sanitization: The instructions do not specify any validation or sanitization steps for the input data before it is passed to other tools.
- [NO_CODE]: The provided skill contains only instructional text and YAML metadata. It does not include any Python scripts, Node.js code, or shell commands, which significantly limits the direct execution risk.
Audit Metadata