radkit-remote-access

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references an external implementation repository at https://github.com/CiscoDevNet/radkit-mcp-server-community. This repository is provided by a well-known technology vendor and is used to deliver the core functionality of the skill.
  • [COMMAND_EXECUTION]: The skill defines tools such as exec_cli_commands_in_device and exec_command for remote device interaction. These capabilities are central to the skill's functionality and include parameters for timeouts and output truncation to maintain control over execution.
  • [PROMPT_INJECTION]: The skill defines an interface for ingesting data from remote network devices (CLI output and SNMP attributes), which represents a vulnerability surface for indirect prompt injection. Ingestion points: exec_cli_commands_in_device and get_device_attributes. Boundary markers: Output truncation is available via max_lines, though explicit instruction-isolation markers are not specified. Capability inventory: Remote CLI and SNMP execution. Sanitization: No explicit sanitization of device-returned content is mentioned. This surface is characteristic of network management tools and no malicious patterns were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 12:32 AM