agentphone
Warn
Audited by Socket on Aug 11, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
Mostly coherent and purpose-aligned telephony API skill with official same-domain data flows and no suspicious installer behavior. The main risk is not malware but the high-impact autonomy: it can place calls, send texts, buy/release numbers, and route conversation data to webhooks, so it is benign in intent but high risk operationally.
Confidence: 89%Severity: 74%
Audit Metadata