add-signal

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This prompt instructs the agent to request and embed a user's phone number, verification code, and captcha token directly into command-line registration/verify commands (e.g., signal-cli -u +NUMBER register / verify CODE), which requires outputting sensitive secret values verbatim and therefore poses an exfiltration risk.

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

• Potentially malicious external URL detected (high risk: 1.00). The install step downloads and installs runtime code from GitHub (curl to https://api.github.com/repos/AsamK/signal-cli/releases/latest and wget https://github.com/AsamK/signal-cli/releases/download/.../signal-cli-${SIGNAL_CLI_VERSION}.tar.gz), which fetches and installs an executable (signal-cli) that the skill runs—i.e., remote code is fetched and executed at runtime.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill instructs the agent to run privileged installation and filesystem operations (e.g., "sudo apt-get install", extracting into /opt and symlinking into /usr/local/bin) and to manage background daemons, which require elevated privileges and therefore can modify the machine state.