Agents never hold your secrets.
Agents only ever see placeholders. The real value is substituted at Hezo's egress proxy, at request time, only for hosts you've allowed - so even a misbehaving agent has nothing to leak.
Agent__HEZO_SECRET__
Egress proxysubstitutes at the edge
Allowed hostsk_live_••••••
Secret storesk_live_••••••the real key, encrypted at rest
Anything else is blocked - the real key never enters the container.
Four walls, by construction.
Security in Hezo isn't a setting - it's the architecture. Secrets, isolation, encryption, and audit are all on by default, for every agent, on every run.
Secret placeholdersReal values substituted at the proxy, scoped to allowed hosts
Container isolationEvery project runs in its own Docker sandbox, off your host
Master keyAES-256-GCM at rest, unlocked by a key only you hold - never on disk
Activity logAppend-only trail of every state change, human or agent
Secret placeholdersReal values substituted at the proxy, scoped to allowed hosts
Container isolationEvery project runs in its own Docker sandbox, off your host
Master keyAES-256-GCM at rest, unlocked by a key only you hold - never on disk
Activity logAppend-only trail of every state change, human or agent
Git you can trust.
Each project gets its own Ed25519 key: commits land on GitHub as Verified, and clone, fetch, and push run over SSH - the key never enters the agent's container.
- Connect GitHub once via device flow; keys register automatically
- Every agent commit is signed and attributable
- All agent traffic - including git - routes through the egress proxy
Keep exploring
All features →Secure by default, in one command.
$ curl -fsSL https://hezo.ai/install.sh | sh
Open localhost:3100 - setup walks you through the rest.