"I built an open source AI agent governance kernel in 3 days as a non-coder"

"Just posted AgentGuard on Hacker News 🛡️
AI agent governance kernel — built in 3 days

Would mean a lot if you checked it out 🙏

You're building with AI agents. Your agent has access to your database, your payments, your users.

What happens when it does something it shouldn't?

Most startups have no answer. No pause button. No audit trail. No trust scoring. No approval gate. Just an agent with full access and a prayer.

AgentGuard changes that.

Mangomindai / agentguard

AI Agent Governance Kernel — trust, approvals, audit

AgentGuard

AgentGuard is a governance kernel for AI agents — trust scoring, capability tokens, human approval gates, and full audit logs

"Your AI agents are only as safe as the system governing them."

---

The Problem

You're building with AI agents. Your agent has access to your database, your payments, your users.

What happens when it does something it shouldn't?

Most startups have no answer. No pause button. No audit trail. No trust scoring. No approval gate. Just an agent with full access and a prayer.

AgentGuard changes that.

---

What AgentGuard Does

Agent wants to take an action
              ↓
AgentGuard checks
  → Is this agent trusted enough?
  → Is this action too risky?
  → Does a human need to approve this?
  → Is this agent even allowed to act right now?
              ↓
ALLOW    → action proceeds, logged
PENDING  → paused, human notified for approval
BLOCK    → stopped, agent trust score drops

…

View on GitHub