Skills
Navigate
skills/automateyournetwork/netclaw/gns3-link-management/Gen Agent Trust Hub

gns3-link-management

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script mcp-servers/gns3-mcp-server/gns3_mcp_server.py using the python3 interpreter to manage the GNS3 bridge.
  • [CREDENTIALS_UNSAFE]: The skill requires the GNS3_PASSWORD environment variable to authenticate with the GNS3 server. While this is a standard configuration for this tool, it involves handling high-value secrets within the execution environment.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through data ingested from the GNS3 API.
  • Ingestion points: The gns3_list_links tool fetches link and interface details from a GNS3 project.
  • Boundary markers: None are defined to separate tool output from agent instructions.
  • Capability inventory: The agent can perform disruptive actions including gns3_delete_link and gns3_isolate_node based on the processed input.
  • Sanitization: No sanitization or validation of the remote project data is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 04:29 PM
Security Audit — agent-trust-hub — gns3-link-management