eve-ng-lab-management

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes external data from the EVE-NG platform (lab metadata, node configurations, and image lists), creating a potential surface for indirect prompt injection.\n
  • Ingestion points: Data enters the agent context through tools such as eve_list_labs, eve_get_lab, and eve_list_images (found in SKILL.md).\n
  • Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions within processed lab data.\n
  • Capability inventory: The skill allows for lab creation, deletion, and archive import/export (SKILL.md), and provides instructions to execute the system binary /usr/sbin/ishare2 (references/operational-guardrails.md).\n
  • Sanitization: No specific sanitization or validation logic for external platform content is described in the provided files.\n- [COMMAND_EXECUTION]: The skill documentation instructs the agent to utilize the /usr/sbin/ishare2 system utility for performing image inventory checks when requested images are missing (references/operational-guardrails.md). While this is a common management utility for EVE-NG environments, it involves the execution of a system binary via an absolute path.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 04:29 PM
Security Audit — agent-trust-hub — eve-ng-lab-management