Open-source governance layer for autonomous AI agents

AI agents that work on your terms.

An open-source trust layer that decides what agents can reach and what they can do — running inside your own environment.

Become a design partner

Ask · check · act · log

The concept

One checkpoint between agents and you.

Security principles

The model, in five parts.

Local by design

Super Tanks runs inside your own environment, so sensitive data can be kept local when using a local model, rather than being sent to a conventional cloud-first service.

Default deny

An agent action that has not been explicitly permitted is refused.

Explicit permissions

Agents can only perform actions that have been explicitly permitted.

Full traceability

Activity is logged, so you can see afterwards what an agent did and on what basis.

Human approval for risky actions

Risky actions can be stopped and held at GO-Gate until a person approves them.

What this enables

Technical properties, practical value.

Use the capability

Bring agents into everyday work without paying with your information.

Keep the data

Sensitive data can be kept local when using a local model.

Govern clearly

Permissions are explicit and written down.

Decide the actions

Anything not permitted is refused by default.

Open source

Open source, built in the open.

1,600+

automated tests

layers

Mapped against

STRIDE
OWASP Agentic Top 10
MITRE ATLAS

Read it yourself

GitHub

Working on AI agents in a sensitive environment?

Become a design partner