AgentGuard by Nano

• Analyzed: May 11, 2026, 5:02 AM
• Findings: 1
• Version: 0.4.0
• Hash: 03c18dd4c341a79b8c8e8c6ac131202aef5b0fadaa1224736c4390856393b7e9
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill is classified as suspicious due to a critical shell injection vulnerability found in `src/1password.js`. User-controlled inputs (such as agent IDs or credential keys) are directly interpolated into `child_process.execSync` calls without proper sanitization or escaping when interacting with the `op` CLI. This flaw, exposed through `src/vault-op.js` and the `agentguard` CLI (`src/cli.js`), could allow an attacker to execute arbitrary commands on the host system. While the skill's stated purpose is security-focused, this severe vulnerability poses a significant risk of remote code execution. Additionally, `src/cli.js` uses a weak default master password ('default-password-change-me') if not provided via environment variables, which is a minor security vulnerability.