Klent is an AI agent guardrail system that intercepts tool calls made by AI agents, allowing developers to allow, deny, modify, steer, or pause each call. It solves the problem of uncontrolled, potentially destructive actions by AI agents, such as unintended database writes or refunds. The product targets developers and IT teams in companies of various sizes, particularly those using AI agents in production. It is positioned as a drop-in solution that works with existing tools and agent frameworks, offering a two-line SDK and a full audit trail.
Key features
- Inline guardrails with policy engine
- Human-in-the-loop approvals
- Action rewrites (mask PII, force dry-run)
- Multi-approver voting
- Audit trail for all tool calls
- Drop-in SDK for TypeScript and Python
- MCP server integration
- Works with Claude Desktop, Cursor
- No social media activity within the last 30 days
GTM channels
- API
- Docs
ICP
- Software developers
- Engineering teams
- IT teams