Free during Alpha
All features included. No credit card required. Sign up and start governing your AI agents.
Everything you need to govern AI agents — included from day one.
- Unlimited agents
- Unlimited evaluations
- Kill switch
- Custom policies + builder
- Approval workflows
- Full audit trail
- Webhooks + CSV export
- PDF compliance reports
- Multi-tenant workspaces
No credit card required. Pricing will be introduced after alpha.
Frequently asked questions
Can't find what you're looking for? Reach out to our team and we'll get back to you.
What is AgentBouncr?
AgentBouncr is an open-core governance framework that sits between your AI agents and the tools they use. It enforces policies, logs every action, and gives you a kill-switch to block further agent actions on demand.
How does the policy engine work?
You define policies via the dashboard or programmatically via the REST API. Specify which tools each agent can access — allow, deny, or require human approval. Policies are evaluated on every tool call before execution. No code deployment needed to change a policy.
Is my audit data tamper-evident?
Yes. Every audit event is linked via a SHA-256 hash chain. Any modification to historical records breaks the chain — you can verify integrity on demand via the audit verification endpoint.
Can I self-host AgentBouncr?
Absolutely. The core is source-available under ELv2. The Enterprise edition adds multi-tenant isolation, OIDC/SAML SSO, Stripe billing, and a managed dashboard.
What AI frameworks are supported?
AgentBouncr is framework-agnostic. It ships native integrations for LangChain and CrewAI (Python) and works with any agent that makes tool calls via HTTP — REST API included.
How does the free tier work?
AgentBouncr is currently in Open Alpha. All features are included for free — unlimited agents, unlimited evaluations, full audit trail, kill switch, and multi-tenant workspaces. No credit card required. No limits. We’ll introduce pricing after the alpha phase.