Terms of use

Last updated 11 November 2026.

The service

Injection Check scans content you send and returns a verdict about whether it contains instructions aimed at an AI agent. By using the API, the MCP endpoint or the demo you agree to these terms.

No guarantee of detection

No scanner catches every attack, and some safe content will be flagged. A SAFE verdict is not a guarantee. An UNKNOWN verdict means the check did not complete and must not be treated as safe. Use Injection Check as one layer of defence, not the only one. You remain responsible for what your agents do.

Acceptable use

• Send only content you are allowed to process.
• Do not try to overload, disrupt or get around the limits of the service.
• Do not use the service to break the law.

Free tier

The free tier allows 100 checks per day per IP address. We may change or withdraw the free tier, and we may block abusive traffic.

Privacy

We never store or log submitted content. See the privacy page.

Warranty and liability

The service is provided “as is”, without warranties of any kind. To the fullest extent the law allows, we are not liable for any loss or damage that results from using the service or relying on its verdicts.

Changes

We may update these terms. The date above shows the latest version.

Contact

hello@injectioncheck.com