aliyun-emoji

Pass

Audited by Gen Agent Trust Hub on Oct 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides integration with official Alibaba Cloud Model Studio Emoji APIs. All referenced URLs in the documentation (e.g., help.aliyun.com) belong to a well-known technology provider and do not present a security risk.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external inputs such as image URLs and template identifiers to generate a JSON request file. This is a standard data processing workflow for an AI agent integration skill.
  • Ingestion points: Parameters including image_url and template_id passed to the scripts/prepare_emoji_request.py utility.
  • Boundary markers: None present in the instructional text.
  • Capability inventory: The skill includes a local script capable of writing JSON request payloads to the file system.
  • Sanitization: The parse_bbox function in the preparation script ensures that bounding box coordinates are strictly validated as integers.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 15, 2026, 08:17 AM