Support and Availability

Operational support expectations for MemoryOS without overpromising a formal SLA.

Last updated: July 24, 2026

MemoryOS aims to provide dependable API and dashboard service. This page describes support expectations. It is not a formal SLA and does not provide service credits unless a signed enterprise agreement says otherwise.

Support Channels

Primary support: [e-mail]

For urgent production incidents, include Production incident in the subject line and provide workspace name, affected endpoint, request IDs, timestamps, and impact.

Availability Approach

MemoryOS is designed around health checks, background job queues, retryable extraction jobs, dead-letter visibility, rate limits, quota controls, provenance, audit logs, and dashboard monitoring.

Maintenance

Maintenance may be performed to deploy fixes, update dependencies, migrate data, or improve reliability. When practical, we will avoid disruptive changes and communicate material production-impacting maintenance.

Incident Handling

For confirmed incidents, MemoryOS may investigate logs, job queues, infrastructure health, third-party dependencies, and affected tenant requests. We may provide status updates where appropriate.

Enterprise Terms

Enterprise customers may request a separate written agreement covering support response targets, availability commitments, data processing, security review, and escalation contacts.