Security audit
Agent Relay
Security checks for vulnerabilities and agentic risk
Overview
The plugin's code, instructions, and configuration align with its stated purpose (waking agents and delivering replies to channel); nothing requests unrelated credentials or performs unexplained network exfiltration.
This plugin appears to do exactly what it says: wake agents in their existing sessions and deliver replies to the user channel. Before installing, ensure you: 1) set a strong authToken for the HTTP /notify endpoint; 2) keep gatewayToken secret (it grants gateway RPC privileges); 3) configure allowedTargets to limit which agents can wake which sessions (omit this only if you want global wake ability); 4) run the gateway on localhost or behind proper network controls (the plugin connects to ws://127.0.0.1:<gatewayPort>); and 5) review logs/usage after enabling because agents (or a compromised agent) could be used to spam users if ACLs are too permissive. If you need higher assurance, ask the maintainer whether device identities are intentionally non-persistent and whether gateway pairing/registry policies should be applied.
Static analysis
No suspicious patterns detected.
