Autonomy where it helps. Control where it matters.
AMVY keeps policy, identity, cost, and audit context attached to every run and consequential action.
Human review
Sensitive tool calls wait for an attributed decision. Auto-approval applies only where your policy allows it.
Identity & access
Single sign-on, SCIM provisioning, role-based permissions, and data isolated per organization.
Guardrails
Sensitive data is screened before it reaches a model, and tool output is checked for prompt injection.
Audit trail
Every run, decision, and applied change is recorded with who did it and when.
Budgets & usage
See spend by team, project, and model, and set the limits each one runs within.
Your data, your rules
Export your organization's data or delete it, under a documented deletion policy.
Run it your way.
Start in the cloud or keep everything inside your own infrastructure, with the models and execution environments you already trust.
AMVY Cloud
Managed by us, with an isolated database for each organization.
Self-hosted
Deploy on your own servers or Kubernetes cluster with Docker, Helm, or Terraform.
Your models
OpenAI, Anthropic, Gemini, Azure OpenAI, AWS Bedrock, or local models through Ollama.
Your execution environments
Agents work in isolated sandboxes on Docker, Kubernetes, AWS microVMs, E2B, Modal, or Vercel.
Give your team and its agents one place to work.
Tell us how your team wants to use AMVY. We review every beta request and reply by email.