Carto – structural intelligence for AI coding agents (OSS)
Blast radius detection before AI edits code, competing with Cursor's codebase awareness.

Human-in-the-loop approval for destructive agent actions before they hit production.
Engineering teams deploying AI agents to production
Oso · Permit.io · OpenPolicyAgent
Blast radius detection before AI edits code, competing with Cursor's codebase awareness.
Deterministic Terraform risk scorer beats AI guessing for deploy safety.
Per-request cryptographic signing fixes MCP's OAuth non-repudiation blind spot.
Walkthrough mode turns code review into pair programming where the agent points at your screen.
Enterprise auth for MCP when the protocol itself has no security layer built in.
TLS for MCP agents with ECDSA passports and L0-L4 trust levels, zero dependencies.