I built a safety shield for AI agents that intercepts dangerous commands

(github.com) by Thomas_Zhou | Jul 6, 2026 | 0 comments on HN