agent-relay
Warn
Audited by Snyk on Oct 5, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The agent relay skill and its associated hooks (
prompt-inbox.shandstop-inbox.sh) poll unread messages from a Relaycast chat/messaging feed where external users or peer agents can submit free text messages.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.70). The skill executes runtime setup and hooks that communicate with
https://api.relaycast.devand download a third-party MCP packageagent-relayvianpx, constituting a weak/unknown/suspicious operational external dependency.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata