cloakify

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill explicitly instructs encoding sensitive files into innocuous-looking strings and pasting those lines as network queries (e.g., DNS hostnames), which requires the agent to produce and output secret-derived values verbatim for exfiltration.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Although hosted on GitHub (a legitimate domain), this repository explicitly provides a data‑exfiltration tool (Cloakify) and instructions to download and run Python scripts that can be readily abused to bypass DLP and exfiltrate sensitive data, so downloading/executing from it is high risk unless from a fully trusted, audited source.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The content explicitly documents a tool whose primary purpose is covert data exfiltration by encoding payloads as innocuous-looking strings (tweets, chess moves, etc.), includes concrete workflows to bypass DLP (DNS exfiltration, noise insertion/removal), and provides decoding on a command-and-control side — indicating clear malicious intent and abuse capability.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's Quick Start explicitly instructs cloning and running code from https://github.com/TryCatchHCF/Cloakify, which fetches remote code that is then executed (e.g., python cloakify.py), so this URL is a runtime dependency that executes remote code.