textme

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The required workflow runs a local daemon that continuously polls inbound iMessages/SMS via Sendblue and feeds them into a Claude Code session as a remote control channel.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill references an external GitHub repository (njerschow/textme) for installation and execution instructions (git clone, npm install), representing an unofficial operational dependency.