What security baseline do you require before adopting an AI workspace?

For teams handling sensitive workflows, what is your minimum security baseline before adopting an AI workspace?

We are building Kachilu for security-first team AI operations (launching March 2026). Design goals:

- per-workspace isolated virtual network boundary

- private east-west networking inside a workspace (agents/tools can communicate)

- secrets encrypted with a workspace key (env vars, volumes, browser cookies)

- per-workspace key management (BYOK planned)

- one-button 24/7 Codex environment + an agent-optimized browser runtime (stealth mode included)

If you can share practical requirements, please pick one:

- workspace isolation model you need

- audit log detail required for incident review

- approval policy for high-impact workflow actions

Short concrete answers are ideal and we will turn feedback into launch updates.

Invite 1 friend to move up in queue. Top referrals can unlock Founding Member status (badge + permanent launch benefit).

5 views