alibabacloud-emr-spark-manage
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: The skill accesses external data by reading Spark job logs, standard output/error files, and SQL execution results from Alibaba Cloud OSS storage using the
ListLogContentsandListSqlStatementContentsAPIs (referenced inreferences/job-management.mdandreferences/api-reference.md). - Boundary markers: The instructions lack explicit delimiters or warnings to treat content from these files as untrusted, which could lead the agent to follow malicious instructions embedded within logs.
- Capability inventory: The skill possesses significant capabilities through the
aliyunCLI, including creating workspaces, submitting jobs, managing Kyuubi services, and modifying RAM permissions (documented inreferences/ram-policies.mdandreferences/workspace-lifecycle.md). - Sanitization: No sanitization or verification of the fetched log/result content is implemented before it is presented to the agent for analysis.
Audit Metadata