Show HN: Scan your AI agents for dangerous capabilities

Category: security

Tags: ai-security, agent-governance, audit-trail

Score: 7.3/10 (Innovation: 7, Technical: 8, Documentation: 8, Utility: 6)

MakerChecker is an open-source security layer for AI agents that enforces deny-by-default permissions, human approvals, and a cryptographically signed audit trail to prevent agents from performing dangerous actions autonomously. It integrates with popular frameworks like LangChain and Claude SDK, and offers a scan tool to identify risky capabilities, making it a timely solution for safe AI agent deployment in regulated industries.

Target audience: backend devs, devops, security engineers

Repository: https://github.com/makerchecker/MakerChecker · TypeScript · AGPL-3.0 · 1 stars

View on Hacker News