Trust & Security

How Kajoo.ai handles oversight, access, and security.

Real answers to the questions your security team will actually ask.

Why agencies run on Kajoo.ai

Security

Human oversight and control

No single point of unsupervised execution.

Every meaningful step has a human checkpoint, and the person approving a change is never the same automated process that executed it.

  • Every tool call can require approval, run automatically, or always-allow, configurable per policy
  • A single, centralized notification and approval center covers every project across your workspace
  • Planning mode: a specialized agent prepares a full plan for review and revision before any execution starts

Data handling

Your data is not training material.

Kajoo.ai does not use customer data to train or fine-tune AI models unless you explicitly agree in writing.

  • Deleted projects are held in a recycle bin for 14 days, then permanently deleted

Access control

Custom roles and granular permissions

Built for enterprise security and governance requirements. Configure oversight once across your whole workspace, or set it project by project.

Infrastructure and security

Kajoo.ai is hosted on Azure and routed through Azure AI Foundry and a Cloudflare secure gateway. Full OAuth support is built in.

  • AI providers and infrastructure partners are named in our Privacy Policy: Microsoft Azure, OpenAI and Anthropic
Read the Privacy Policy

Audit and observability

A full record of what happened, and when.

Every job is tracked on an auto-updating kanban task board with a complete audit trail.

Questions your security team hasn't seen answered here?

Talk to us directly