Terms of Service
Last updated: May 2026
1. Acceptance of Terms
By accessing or using Sentinel Proxy ("the Service"), you agree to be bound by these Terms of Service and our Privacy Policy. If you do not agree, do not use the Service.
2. Service Description
Sentinel Proxy is an AI Firewall (AIF) that inspects prompts sent to large language models (LLMs) for prompt injection attacks. The Service operates as a reverse proxy between your application and your upstream LLM provider, scanning requests in real time and taking action (clean, neutralize, or block) based on detected threat levels.
3. Data Processing and Retention
By default, Sentinel Proxy does not store, log, or persist the content of your prompts or LLM responses.
All prompt content is processed entirely in memory for threat analysis and discarded immediately after the request completes. We retain only aggregate usage metadata (request counts, threat scores, action taken, and latency) for your dashboard analytics. See our Privacy Policy for full details.
If you enable the optional Community Training Programme (see Section 4), flagged or blocked prompt content may be temporarily stored for admin review and signature improvement. This feature is disabled by default and requires your explicit consent. You may withdraw consent at any time by toggling off Community Training in your account Settings.
4. Community Training Programme
The Community Training Programme is a voluntary, opt-in feature available to subscribers on eligible plans. When enabled:
- Content of flagged or blocked prompts is stored in a secure review queue
- An administrator may review, approve, or reject entries
- Approved entries are used to generate anonymised detection signatures that improve Sentinel’s security engine
- Clean prompts are never captured under any circumstances
By enabling this feature, you grant Skyblue Soft a non-exclusive, royalty-free licence to use the stored content solely for the purpose of improving the threat detection engine. This licence terminates upon opt-out and any pending (unreviewed) entries are deleted immediately. You remain responsible for ensuring you have authority to share any content that flows through your integration.
5. Acceptable Use
You agree not to:
- Use the Service for any unlawful purpose or to facilitate illegal activity
- Attempt to bypass, disable, or interfere with the Service's security mechanisms
- Reverse engineer, decompile, or attempt to extract the source code of the threat detection engine
- Exceed your plan's rate limits through automated means or abuse the Service infrastructure
- Resell or redistribute access to the Service without written authorisation
- Submit content through the Community Training Programme that you do not have the right to share
6. API Keys and Account Security
You are responsible for safeguarding your API keys. API keys are shown only once at creation and are stored as irreversible SHA-256 hashes. Treat your keys as secrets—do not commit them to public repositories or share them in plaintext. You are responsible for all activity that occurs under your API keys. Notify us immediately if you believe a key has been compromised.
7. Subscription and Billing
Paid plans are billed monthly through Stripe. You may cancel or change your plan at any time through the billing portal on your settings page. Downgrades take effect at the end of the current billing period. No refunds are provided for partial months. The free Starter tier requires no credit card and has no time limit.
8. Service Availability
The Service is provided on an "as is" and "as available" basis. We make commercially reasonable efforts to maintain uptime but do not guarantee uninterrupted availability. We are not liable for any losses caused by service downtime, latency, or false positives/negatives in threat detection.
9. Limitation of Liability
To the maximum extent permitted by law, Sentinel Proxy and its operators shall not be liable for any indirect, incidental, special, consequential, or punitive damages, including but not limited to loss of profits, data, or business opportunities, arising from your use of or inability to use the Service. Our total liability for any claim shall not exceed the amount you paid us in the twelve months preceding the claim.
10. Termination
We may suspend or terminate your access to the Service at any time for violation of these terms, abuse of the platform, or non-payment. You may terminate your account at any time by cancelling your subscription and deleting your API keys. Upon termination, your usage data will be deleted in accordance with our retention schedule, and any pending Community Training queue entries will be deleted immediately.
11. Governing Law
These terms are governed by and construed in accordance with applicable law. Nothing in these terms limits your statutory rights under the GDPR, UK GDPR, or other applicable consumer protection legislation. For GDPR enquiries, see our Privacy Policy.
12. Changes to Terms
We may update these terms from time to time. Material changes will be communicated via the email associated with your account. Continued use of the Service after changes take effect constitutes acceptance of the updated terms.
13. Contact
Questions about these terms? Reach us at support@skyblue-soft.com.