alibabacloud-cadt-arch-draw

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's documentation provides instructions to download and execute an installation script for the aliyun CLI from the official vendor domain (aliyuncli.alicdn.com).
  • [COMMAND_EXECUTION]: The skill operates by executing aliyun CLI commands to perform cloud operations, specifically using the bpstudio service for architecture generation.
  • [PROMPT_INJECTION]: The skill acts as a relay for user-provided architecture descriptions, which are passed directly to a backend AI service. This represents an indirect prompt injection surface.
  • Ingestion points: User-provided resource descriptions (e.g., "一台 Web ECS + 一台 MySQL RDS") enter the agent context during Step 1 of the flow in SKILL.md.
  • Boundary markers: User input is encapsulated within the JSON string provided to the --attributes parameter of the CLI command.
  • Capability inventory: The skill executes aliyun bpstudio execute-operation-sync, which triggers AI-driven architecture creation and validation on the cloud backend (documented in SKILL.md and references/related-commands.md).
  • Sanitization: The skill relies on the Alibaba Cloud backend to validate and sanitize the input strings provided in the prompt attribute.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 02:21 AM
Security Audit — agent-trust-hub — alibabacloud-cadt-arch-draw