fynd-backend-microservices
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The script
scripts/diagnose.jsuseschild_process.execSyncto executekubectl get pods. This is a direct shell command execution pattern. While the command is static and aligned with the primary diagnostic purpose of the skill, it utilizes a sensitive capability that grants the agent access to infrastructure metadata. - [INDIRECT_PROMPT_INJECTION] (LOW): The skill exhibits an attack surface for indirect prompt injection due to its data ingestion and execution capabilities.
- Ingestion points: Data enters the agent context through the output of the
kubectlcommand inscripts/diagnose.js. - Boundary markers: Absent. The script does not wrap the environment output in delimiters or provide instructions to ignore embedded commands.
- Capability inventory: The skill uses
execSyncinscripts/diagnose.jsfor command execution. - Sanitization: Absent. External data from the cluster (such as maliciously named pods) is not sanitized before being returned to the agent context.
Audit Metadata