aliyun-qwen-text-embedding
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The SKILL.md file includes standard shell commands to create output directories, validate Python script compilation, and run the embedding request preparation script. These operations are limited to the local environment and are intended for validation and execution of the included tool.\n- [INDIRECT_PROMPT_INJECTION]: The script
prepare_embedding_request.pyaccepts arbitrary user text via the--textargument. While it correctly uses JSON serialization to prevent payload corruption, it represents an ingestion point for untrusted data that will be processed by downstream embedding models. The script includes no dangerous capabilities such as network exfiltration or subprocess execution.\n- [EXTERNAL_DOWNLOADS]: The references file links to official Alibaba Cloud (aliyun.com) documentation for Model Studio embedding services. These are well-known technology service domains and are used neutrally to provide model specifications and API information.
Audit Metadata