Qwen Code

• Analyzed: May 11, 2026, 4:50 AM
• Findings: 1
• Version: 1.2.0
• Hash: 2d83b61a8c3073aedb4c868e2393e2772cba60191973e8d41e9c9bfa8d3413e2
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill is a wrapper for the powerful 'qwen' AI coding CLI, which inherently carries high-risk capabilities such as generating, executing, and modifying code, and potentially reading sensitive files (as explicitly warned in SKILL.md regarding `~/clawd/`). The script `scripts/qwen-code.js` uses `child_process.spawn` to execute `qwen` commands with user-controlled prompts, which could facilitate prompt injection against the `qwen` CLI itself, leading to arbitrary code execution or unauthorized data access. While the skill's documentation attempts to mitigate these risks by warning about 'YOLO' (auto-approve) mode and sensitive directories, the underlying capabilities are significant vulnerabilities without clear malicious intent from the skill's author.