Show HN: Hashling, an open-source Mac security agent your own AI can talk to (github.com)

Hashling has been introduced as an open-source security agent designed specifically for Mac users, allowing personal AI assistants like Claude, Cursor, or Codex to enhance cybersecurity. This tool continuously monitors for scams, malware, and any unusual behaviors from AI applications running on the Mac. The standout feature is its permission-based approach, where changes are only executed with user approval via Touch ID. Hashling simplifies security checks, offering detailed insights about permissions and suspicious activities, while maintaining user privacy by keeping all data local. The significance of Hashling lies in its integration with AI technologies, making it an essential tool for users leveraging AI agents for various tasks. As AI-generated scams become increasingly common, Hashling equips users with the ability to validate messages and links before interaction. Built on open-source foundations, it includes a variety of checks ranging from malware detection to assessing the safety of AI applications, all while prioritizing user consent. By offering a user-friendly interface and detailed explanations, Hashling aims to make cybersecurity accessible, ensuring users can protect their devices without extensive technical knowledge.