Terms of Service
By accessing or using the Neusnap API, documentation, or related services, you agree to these Terms of Service.
Neusnap provides infrastructure for routing, execution, cost control, and state management of AI agent workflows. The service is provided on an "as-is" and "as-available" basis.
Neusnap does not guarantee uninterrupted availability, specific model behavior, or compatibility with any third-party provider. Model outputs, latency, and pricing characteristics are ultimately determined by underlying model providers.
Scope of Responsibility
You are responsible for:
- The correctness and safety of your application code and business logic
- How outputs generated by models are interpreted or used
- Compliance with applicable laws, regulations, and internal policies
Neusnap Guarantees vs Limitations
Neusnap provides execution-level guarantees, not semantic or factual guarantees.
Neusnap is not responsible for:
- The factual accuracy, completeness, or semantic correctness of model-generated content
- Business, legal, or operational decisions made using model outputs
- Third-party service outages, provider-level failures, or upstream API behavior
- Losses caused by misuse or misconfiguration of the API
Clarification on Model Outputs
Neusnap includes infrastructure mechanisms to detect, constrain, retry, and recover from malformed, failed, or inconsistent model responses, including rollback-safe execution and deterministic retry behavior. However, Neusnap does not guarantee that any model output is factually correct, truthful, or suitable for a specific purpose.
Neusnap enforces structural correctness and execution guarantees, not semantic intent.
Service Changes
Neusnap may modify, suspend, or discontinue the service at any time, with reasonable efforts to communicate material changes.
Use of the service constitutes acceptance of these terms.
For questions regarding these terms, contact support@neusnap.net.