wechat-cli-local-data

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). Mixed: the apple.com DTD is legitimate, but the presence of a GitHub repo clone (https://github.com/freestylefly/wechat-cli.git) and an unknown personal domain (https://ara.so), combined with the tool's install/run instructions that require elevated privileges, process-memory scanning and re‑signing of apps, are high-risk indicators for potential misuse or malware distribution (the bare "https://" is malformed); treat the repo/personal domain as suspicious until vetted.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill intentionally scans process memory to extract SQLCipher encryption keys, stores them locally, requires root/elevated privileges and even re-signs the WeChat binary (get-task-allow), which are deliberate actions enabling credential theft and system compromise and make the tool high-risk for abuse (especially when integrated into automated agents), even though no explicit network exfiltration is shown.

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 explicitly instructs running commands with sudo (e.g., sudo wechat-cli init) and even re-signing the WeChat app via sudo codesign to change entitlements and enable memory scanning, which requires privilege escalation and modifies system application files.