aliyun-ecs-manage

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a script scripts/run_remote_command.py that utilizes the Alibaba Cloud 'RunCommand' API to execute arbitrary shell commands on managed ECS instances. This is a core functionality of the skill intended for remote administration and troubleshooting.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of official Alibaba Cloud SDK packages (e.g., alibabacloud_ecs20140526, alibabacloud_cms20190101) from standard package registries. These are recognized dependencies for interacting with Alibaba Cloud services.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing data from external API responses, such as instance metadata and command output from remote servers.
  • Ingestion points: API responses processed in scripts/list_instances_all_regions.py and command output retrieved in scripts/run_remote_command.py.
  • Boundary markers: None identified in the script outputs or documentation instructions.
  • Capability inventory: The skill can perform remote command execution on cloud instances and write data to the local output/ directory.
  • Sanitization: The skill decodes base64 output from the cloud assistant but does not perform additional sanitization or filtering on the retrieved content before displaying it.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:20 AM