Janee

• Analyzed: May 12, 2026, 3:21 AM
• Findings: 1
• Version: 0.1.2
• Hash: a74543eb180507ee0abbf202675eafcd66bfa0e0c0587658a1821d0e4957b46e
• Source: palm
• External report: View on VirusTotal

Findings (1)

The OpenClaw AgentSkills skill bundle for Janee is designed as a security tool for AI agents, focusing on secrets management and API access control. It encrypts API keys at rest, prevents agents from directly accessing keys, enforces path-based access policies, and maintains a full audit log of all API requests. The code demonstrates strong security practices, including the use of AES-256-GCM encryption for secrets, restrictive file permissions (0o600/0o700) for configuration and logs in `~/.janee/`, and safe URL construction to prevent injection. The `SKILL.md` and `README.md` documentation clearly outlines these security features and does not contain any instructions that could lead to prompt injection or malicious agent behavior. All network and file access is directly aligned with its stated purpose of securely proxying API calls.