pyats-f5-platform
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes Python scripts via the MCP framework to execute pyATS commands on F5 network devices using established testbed configurations.
- [PROMPT_INJECTION]: The skill's design was evaluated for indirect prompt injection risks.
- Ingestion points: Data enters the agent context from external F5 iControl REST API responses (e.g., /mgmt/tm/sys/version, /mgmt/tm/net/interface).
- Boundary markers: Not explicitly defined in the provided markdown.
- Capability inventory: The skill is configured to execute read-only pyATS show commands via the system-defined $PYATS_MCP_SCRIPT.
- Sanitization: No explicit sanitization of API responses is defined in the markdown, as the data is retrieved from trusted infrastructure for monitoring purposes.
Audit Metadata