import
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use standard diagnostic utilities including
curl,grep,jstat,netstat, andtcpdump. These tools are used to query StarRocks cluster metrics from local endpoints (e.g., ports 8040, 8060) and to inspect system logs for performance bottlenecks. - [INDIRECT_PROMPT_INJECTION]: The skill involves reading and analyzing external data sources such as database logs (
fe.log,be.INFO) and metrics output. While these files represent an ingestion surface for untrusted data that could theoretically contain malicious instructions, the skill uses standard filtering (e.g.,grepfor specific txn IDs) which limits the execution surface. This is a standard risk profile for diagnostic and log-analysis agents.
Audit Metadata