Kimi Code CLI

• Analyzed: May 12, 2026, 4:20 AM
• Findings: 1
• Version: 1.0.0
• Hash: ebd147939063c9d58c2f167ff4f091780f6540550b7cfeeedfa0af96f99ce3a7
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill bundle is classified as suspicious due to a critical shell injection vulnerability found in `scripts/kimi_runner.py`. The `task` argument, which can be user-controlled, is directly interpolated into the `command` string passed to `bash` without proper sanitization (e.g., `command:"kimi '{task}'"`). This allows an attacker to break out of the single quotes and execute arbitrary commands on the host system. While the `SKILL.md` itself is not malicious, it describes a powerful skill that grants the OpenClaw agent extensive system access, including arbitrary shell command execution and directory creation, which amplifies the risk of this vulnerability.