Skills
Navigate
skills/mystenlabs/skills/sui-object-model/Gen Agent Trust Hub

sui-object-model

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill recommends querying the Sui documentation MCP server at https://sui.mcp.kapa.ai for real-time answers. This establishes an indirect prompt injection surface where the agent processes content from an external source at runtime. 1. Ingestion point: https://sui.mcp.kapa.ai (MCP tool endpoint). 2. Boundary markers: None. 3. Capability inventory: The skill includes Move documentation and examples but no executable scripts; the agent's base tools constitute the capability set. 4. Sanitization: None.
  • [EXTERNAL_DOWNLOADS]: The skill references established documentation sources including docs.sui.io and move-book.com, and utilizes a well-known documentation search service at kapa.ai. These resources are standard for the developer use-case and are affiliated with the vendor or reputable services.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 04:24 AM