troubleshooting-application-failures
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Indirect Prompt Injection Surface: The skill processes external data from CloudWatch logs, which constitutes a surface for indirect prompt injection if the logs contain instructions designed to influence the agent.
- Ingestion points: Log data enters the agent context through the execution of
aws logs get-query-resultsas described inreferences/application-failure-troubleshooting.md. - Boundary markers: The instructions do not specify explicit delimiters or instructions to the model to ignore potential commands embedded within the retrieved log messages.
- Capability inventory: The agent has the capability to execute shell-based AWS CLI commands through the
call_awstool. - Sanitization: The skill does not explicitly define sanitization or filtering steps for the log content prior to its analysis by the model.
- Command Execution: The skill instructs the agent to use the
call_awstool to execute AWS CLI commands likedescribe-log-groupsandstart-query. This is a core component of the skill's troubleshooting functionality and relies on the permissions associated with the agent's AWS credentials.
Audit Metadata