Yours, 100%. You own everything you store. We take no license to it and cannot read it.
Keep it, and your recovery key, safe. Lose both and nobody — including us — can recover your data. That is by design.
Private beta. We may change, pause or end features while we harden the service. We will give notice where we can.
These terms are an agreement between you and [LEGAL ENTITY NAME], an Ohio limited liability company ("aiSecurityWarden", "we", "us"). By creating an account, using our client software, or using aisecuritywarden.com (together, the "Service"), you agree to them. If you use the Service for an organization, you confirm you may bind it. If you do not agree, please do not use the Service. You must be 18 or older.
aiSecurityWarden is a hosted secrets manager for AI agents and automation. It stores secrets encrypted end to end, issues machine tokens with per-token limits (project, allowed addresses, expiry), can require a person to approve access to secrets and individual agent actions, and delivers connectors that run on your own machines. It includes a website and client software for Windows, macOS and Linux. It is currently a private beta, free during the beta unless we tell you otherwise.
You own everything you store — secrets, connectors, project and token settings, logs. You grant us only the permission needed to store, process and deliver it to run the Service under our Privacy Policy. Because of end-to-end encryption, we handle your secrets and connectors only in a form we cannot decrypt.
The rules you set and the decisions you and your approvers make are your actions, not ours. An action your rules allow without asking (for example "allow" or an open time window) counts as approved by you. We deliver requests, check match codes and timeouts, enforce your rules and record the outcome. We are not responsible for what an agent does with access you allowed, or for the content of messages an agent sends after approval. Review approval requests before approving them.
You agree not to:
Notification channels (Telegram, ntfy), connected apps (such as Gmail and Google Drive), AI agents and AI model providers, and our hosting providers are third-party services governed by their own terms and privacy policies. We are not responsible for them.
If you send us ideas or bug reports, we may use them without obligation. Please don't include secrets or confidential information.
During the beta there is no uptime guarantee. We aim to keep the Service running and to announce planned changes in advance. We may change, suspend or end the beta; before ending the Service we will give reasonable notice and a window to take your data with you.
You may stop using the Service at any time and ask us to delete your account (handled as described in the Privacy Policy). We may suspend or end access that breaks these terms or threatens the Service; except for abuse, we give notice and a chance to take your data with you where reasonable. Sections 4, 5, 14, 15, 16 and 17 survive termination.
The Service is provided "as is" and "as available", to the extent the law allows. We disclaim all warranties, express or implied, including merchantability, fitness for a particular purpose, title and non-infringement. We do not promise the Service will be uninterrupted or error-free, or that it will prevent every security incident or every unwanted agent action. Security tools reduce risk; they do not remove it.
To the extent the law allows, we are not liable for indirect, incidental, special, consequential, exemplary or punitive damages, or for lost profits, lost data, lost secrets or business interruption, even if told they were possible. Our total liability under these terms is limited to the greater of (a) the fees you paid us in the 12 months before the claim arose, or (b) US$100. These limits do not apply where the law does not allow them.
If a third party makes a claim against us because of your use of the Service, your data or connectors, or your breach of these terms, you will defend and compensate us for the resulting costs and damages, except to the extent the claim is caused by our own breach or misconduct.
Questions about these terms, sales and plans, or security reports: use the contact form.