Service Level Agreement
Last updated: October 2026 · Sugra Systems, Inc. · Delaware, USA
What changed in this version. The Sugra MCP Gateway is now covered alongside the Sugra API, the Personal plan has its own monthly target of 98.0%, requests that receive no response now count toward Downtime, and Pro support requests are handled first within each severity. The 99.0% target for every other paid plan is unchanged.
This Service Level Agreement ("SLA") is part of and incorporated by reference into the Sugra Terms of Service. Capitalized terms used but not defined here have the meanings given in the Terms of Service.
1. Scope
This SLA sets out the availability targets for the Sugra API and the Sugra MCP Gateway (each a "Covered Service") and the Service Credits available to paid Subscribers in the event the applicable target is not met. This SLA applies only to paid subscription tiers. Free tiers, evaluation access, beta features, and Enterprise tiers governed by a separate written agreement are not covered by this SLA.
"Sugra MCP Gateway" means the hosted Model Context Protocol (MCP) endpoint that Sugra operates so that AI assistants and agents can use the Service, at any address Sugra operates for that purpose, including an address used by a Sugra listing in a third-party connector directory. Sugra publishes the address to use at https://mcp.sugra.ai and may change it. The Sugra MCP Gateway does not include software that Sugra distributes for the Subscriber to run on the Subscriber's own systems.
Subscribers on Enterprise tiers may negotiate higher availability targets, custom response times, or tailored remedies in a separate written agreement. In the event of conflict, such separate agreement prevails.
The Sugra Earth hazards endpoints (/api/v2/hazards), including when reached through the Sugra MCP Gateway, are excluded from the availability and accuracy commitments of this Service Level Agreement. Hazard data is a best-effort relay of third-party authoritative sources; values are preliminary and subject to revision, and upstream feeds may be delayed or unavailable. No uptime, latency, or accuracy warranty applies to the hazards endpoints.
The Sugra Earth marine endpoints (/api/v2/transport/marine), including when reached through the Sugra MCP Gateway, are likewise excluded from the availability and accuracy commitments of this Service Level Agreement. Marine data is best-effort model output and is not navigation-grade; no uptime, latency, or accuracy warranty applies.
2. Service Availability Target
Sugra targets the following monthly availability of the Covered Services, measured as described in Section 3:
- Personal plan: ninety-eight point zero percent (98.0%).
- Every other paid plan: ninety-nine point zero percent (99.0%).
If a Subscriber held more than one plan during a calendar month, the higher target applies to that month.
3. Uptime Calculation
Monthly availability is calculated as:
Availability (%) = ((Total Minutes in the Calendar Month - Downtime Minutes) / Total Minutes in the Calendar Month) × 100
"Downtime" means a continuous period during which the production endpoints of a Covered Service return server-side errors (HTTP 5xx), or return no response at all, for a majority of valid, authenticated requests originating from unaffected regions and networks, as measured by Sugra's monitoring. A minute in which both Covered Services are in Downtime counts as one Downtime minute.
The following are excluded from Downtime and do not count against the availability target:
- Scheduled maintenance announced in accordance with Section 7.
- Unavailability caused by force majeure, including acts of God, natural disasters, war, terrorism, civil unrest, labor disputes, government action, pandemics, and failures of the public Internet, public cloud infrastructure, or domain name resolution outside Sugra's immediate control.
- Denial-of-service attacks, distributed denial-of-service attacks, and other malicious traffic originating outside Sugra.
- Client-side issues, including misconfigured requests, expired or revoked credentials, exhausted rate limits or quotas, and incompatible client implementations.
- Issues caused by the Subscriber's or a third party's modifications to or integration with the Service.
- Upstream Source unavailability, as described in Section 4.
- Any period during which the Subscriber's account is suspended for breach of the Terms of Service, the Data Use Policy, or the Acceptable Use Policy.
- Unavailability, errors, or limits of a third-party AI assistant, agent platform, or other client used to connect to the Sugra MCP Gateway, including the third party's own sign-in steps.
4. Upstream Source Exclusions
The Service aggregates Data from third-party Upstream Sources. Availability of Data from any particular Upstream Source, as well as the accuracy, completeness, and timeliness of such Data, is not covered by this SLA. If an Upstream Source is unavailable, slow, or returns invalid data, and Sugra is unable to satisfy a given request notwithstanding reasonable automatic failover to alternative sources where available, such unavailability does not constitute Downtime of a Covered Service.
A Covered Service is considered available for purposes of this SLA when it is capable of receiving, authenticating, and responding to requests, even where the response indicates that a particular Upstream Source is unavailable.
5. Service Credits
Where monthly availability falls below the target that applies to a Subscriber under Section 2, the affected Subscriber is entitled to a Service Credit calculated as a percentage of the monthly fees paid by that Subscriber for the Service in the affected month:
- Less than the applicable target and 95.0% or greater: 10% of monthly fees
- Less than 95.0% and 90.0% or greater: 25% of monthly fees
- Less than 90.0%: 50% of monthly fees
Service Credits are applied to future Fees due on the Subscriber's account, do not accrue interest, and are not convertible to cash or equivalent compensation.
Service Credits are the Subscriber's sole and exclusive remedy for any failure of the Service to meet the availability target. The total Service Credits available to a Subscriber in any calendar month shall not exceed fifty percent (50%) of the monthly fees paid by that Subscriber for that month.
6. Incident Response
Sugra aims to respond to operational incidents in accordance with the following target response times, measured from the time Sugra acknowledges the incident:
- Critical - complete unavailability of a Covered Service for paid Subscribers. Target response: four (4) hours.
- Major - substantial degradation of a Covered Service affecting many Subscribers, such as sustained elevated error rates or severely increased latency. Target response: eight (8) hours.
- Minor - isolated issues affecting a specific endpoint, feature, or subset of Subscribers. Target response: two (2) business days.
"Response" means acknowledgement of the incident and initiation of investigation. It does not mean resolution. Sugra will use commercially reasonable efforts to restore the Service promptly. Response times are targets, not warranties, and a failure to meet a target response time does not, by itself, trigger Service Credits.
Support requests from Subscribers on the Pro plan are handled ahead of other requests of the same severity. This priority changes the order of handling, not the target response times above.
7. Scheduled Maintenance
Sugra may perform scheduled maintenance to upgrade, patch, or reconfigure the Service. Sugra will announce scheduled maintenance at least forty-eight (48) hours in advance via email to active Subscribers and a notice at status.sugra.ai. The preferred maintenance window is Sunday, 02:00 to 06:00 UTC. Sugra may perform emergency maintenance without prior notice where reasonably necessary to address security or stability issues.
8. Status Page
Current operational status of the Service is published at status.sugra.ai. Sugra may move or replace this location; the current location will be referenced at sugra.systems. The status page is the authoritative record of incidents and scheduled maintenance for purposes of this SLA.
9. Credit Request Procedure
To receive a Service Credit, the Subscriber must submit a request to support@sugra.systems no later than thirty (30) days after the end of the calendar month in which the alleged unavailability occurred. The request must include:
- The affected account identifier.
- The date range of the alleged unavailability.
- A description of the impact and, where available, relevant request logs, error responses, or similar evidence.
Sugra will review the request and, where the claim is validated against Sugra's monitoring records, apply the applicable Service Credit to the Subscriber's account. Sugra's monitoring records are the primary evidence of availability for purposes of this SLA.
10. Limitations
This SLA is conditional on the Subscriber's compliance with the Terms of Service, the Data Use Policy, and the Acceptable Use Policy. No Service Credits will be issued for any period during which the Subscriber is in material breach of any such document, nor for any period attributable to an excluded event under Section 3 or Section 4.
This SLA does not apply to, and Sugra makes no availability commitment with respect to, the Data itself, including its accuracy, completeness, currency, or fitness for any purpose, all of which are governed by the Data Use Policy.
Sugra may modify this SLA from time to time. For material changes, Sugra will provide at least thirty (30) days' advance notice by email to active Subscribers and a prominent notice at sugra.systems, and will update the "Last updated" date above.
Contact. SLA inquiries and Service Credit requests: support@sugra.systems. Legal inquiries: legal@sugra.systems.