eve-lab-topology-build
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external EVE-NG lab environments, which presents a standard attack surface for indirect prompt injection if the remote data contains malicious instructions.
- Ingestion points: The tools
eve_get_topology,eve_list_networks, andeve_list_node_interfacesingest data from an external EVE-NG API. - Boundary markers: No explicit delimiters or boundary markers are defined in the instructions to isolate retrieved topology data from the agent's prompt context.
- Capability inventory: The skill possesses write capabilities including
eve_create_network,eve_delete_network, andeve_connect_interfacewhich could be targeted by injected instructions. - Sanitization: The instructions do not specify any sanitization or validation procedures for the data returned by the EVE-NG server.
Audit Metadata