aliyun-cli
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external Alibaba Cloud APIs through various monitoring and management commands.
- Ingestion points: The agent processes output from commands such as
aliyun ecs DescribeInstancesandaliyun sls GetLogs(inSKILL.md). - Boundary markers: The instructions do not define delimiters or specific warnings to ignore embedded instructions within the data retrieved from the cloud.
- Capability inventory: The skill provides extensive capabilities for resource creation, deletion, and configuration across ECS, OSS, RDS, and SLB services.
- Sanitization: There is no evidence of sanitization or validation logic for the content returned by the CLI tools.
- Risk: If an attacker can control cloud resource metadata (e.g., instance names) or log content, they could potentially influence the agent's behavior via a prompt injection payload embedded in that data.
Audit Metadata