Show HN: Keydris checks your AI agent's permissions before it sends email

Category: security

Tags: ai-agents, permissions, guardrails, email, llm-security

Score: 5.0/10 (Innovation: 6, Technical: 4, Documentation: 2, Utility: 5)

Keydris is a permission-control layer for AI agents that gates sensitive actions such as sending email. The project targets an emerging need for safety guardrails around autonomous LLM agents, but the submission is only a YouTube video with no code or substantive documentation available for evaluation.

Target audience: backend devs and AI engineers building autonomous agents

Repository: https://www.youtube.com/watch?v=De0PQTEL8r0

View on Hacker News