add-rtk

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). The curl | sh command fetches and executes a remote install.sh from a third-party GitHub repo — executing an unreviewed shell script (even from raw.githubusercontent.com) is high-risk because it runs arbitrary code and can be used to distribute malware.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This repo instructs installing an unsigned third‑party CLI via curl|sh, mounting the host binary into agent containers and adding a PreToolUse hook that transparently routes every Bash command through that CLI — enabling supply‑chain risk, broad command interception, and potential data exfiltration/backdoor behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill runs a remote install script with curl | sh (https://raw.githubusercontent.com/rtk-ai/rtk/refs/heads/master/install.sh), which fetches and executes remote code and installs a binary that the PreToolUse hook relies on to intercept/execute agent commands.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). This skill directs installing a remote script via curl|sh and programmatically changing container configurations and the agents' settings.json to mount a host binary and add a PreToolUse hook, which modifies host/container state and the agent runtime behavior.