eve-ng-node-operations

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Implements several tools for node lifecycle management including creation, deletion, and state control (start/stop/wipe) within an EVE-NG server.
  • [CREDENTIALS_UNSAFE]: The skill metadata requests environment variables for EVE_USER and EVE_PASSWORD to authenticate with the EVE-NG API. This is a standard requirement for such tools and does not indicate hardcoded credentials.
  • [PROMPT_INJECTION]: The skill ingests data from the EVE-NG API and potentially external inventory tools like ishare2. This constitutes an indirect prompt injection surface if strings within the EVE-NG environment (such as node names or descriptions) are manipulated by an attacker.
  • Ingestion points: Responses from eve_list_nodes, eve_get_node, and ishare2 inventory output.
  • Boundary markers: Not explicitly defined in the provided instruction files.
  • Capability inventory: Full lifecycle control of lab nodes including destructive actions like eve_wipe_node and eve_delete_node.
  • Sanitization: No explicit content filtering or escaping is specified for the ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 04:53 AM
Security Audit — agent-trust-hub — eve-ng-node-operations