pyats-health-check
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell command templates to execute pyATS and NetBox tools via python scripts. These commands rely on environment variables such as $PYATS_MCP_SCRIPT and $NETBOX_MCP_SCRIPT for tool locations, which is a standard pattern for this integration.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from managed network devices, which could contain adversarial content designed to influence the agent's logic during processing.
- Ingestion points: Device command outputs from 'show version', 'show processes', and system logs, as well as operational state data from NetBox.
- Boundary markers: None identified; the instructions do not specify using delimiters or ignore-instructions for the external data being parsed.
- Capability inventory: The skill can perform network reachability tests (ping), execute arbitrary show commands on devices, and record session data into the GAIT audit trail.
- Sanitization: Absent; there is no explicit instruction to sanitize or escape device-provided strings before the agent evaluates them against health thresholds or log patterns.
Audit Metadata