cisco-yang-cli

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute arbitrary CLI commands on remote network devices using the cisco-yang exec command, which allows for broad administrative control.
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of data from external Cisco devices, which may contain attacker-controlled content if a device is compromised or configured with malicious strings (e.g., in descriptions or metadata). This constitutes an indirect prompt injection surface.
  • Ingestion points: Data returned from cisco-yang get and cisco-yang exec commands as described in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat device output as untrusted data or to ignore embedded instructions.
  • Capability inventory: The skill provides access to destructive and administrative tools including cisco-yang set, cisco-yang delete, and unrestricted cisco-yang exec.
  • Sanitization: No evidence of sanitization, filtering, or validation of the data received from the network devices before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 05:15 PM