Back to skill
Skillv1.0.0
VirusTotal security
「自己」微信画像分析 · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
SuspiciousApr 3, 2026, 4:06 AM
- Hash
- 99dabdbb15884a247cab09e2f77e8f5230d603ab88d66969c77716ff2072ac64
- Source
- palm
- Verdict
- suspicious
- Code Insight
- Type: OpenClaw Skill Name: ziji-ai Version: 1.0.0 This skill bundle is designed to decrypt and analyze a user's local WeChat database to generate a detailed 'personality portrait,' including financial transactions, relationship networks, and psychological profiles. It is classified as suspicious due to its requirement for Administrator privileges to dump process memory for decryption keys and its use of an external 'setup.py' script that clones unverified code from GitHub (wechat-decrypt). Additionally, it implements a persistence and automation layer using shell scripts and cron jobs that inject instructions into the AI agent's workflow via a HEARTBEAT.md file, creating a high-risk environment for sensitive data handling and potential supply chain attacks.
- External report
- View on VirusTotal