Maybe Don't
Monitors, blocks, and coaches AI agents via a natural language policy engine with audit logging to prevent risky actions
The product is an AI guardrail and policy engine that monitors, blocks, and coaches AI agents to prevent risky or non-compliant actions such as deleting production data, with observability and a comprehensive audit log. It sells to enterprise buyers including developers, engineering teams, and security/CISO stakeholders deploying agentic AI in business environments. It is delivered as open source software under Apache 2.0 that can be self-hosted for free, with SaaS options including free tier and enterprise plans.
Key features
- Monitoring for AI agent actions
- Intelligent blocking of risky actions
- Coaching with explanations for denials
- Natural language policy engine
- Comprehensive audit log
- Agentic observability dashboard
- Governance and compliance controls
- Policy validation and intercept
- CEL policy support
- SIEM integration
- No social media activity within the last 30 days
GTM channels
- Blog
- API
- Docs
ICP
- Engineering teams
- Security teams
- Enterprises