Service Level Agreement
Last updated:
This Service Level Agreement describes Driftless’s uptime commitment, how uptime is measured, support response times by severity level, how we notify you of incidents, and the exclusions that apply.
Uptime commitment
Driftless is an agentic product delivery platform provided by Driftless (“Driftless”, “we”, “us”, “our”). This Service Level Agreement (“SLA”) governs the availability of the Driftless application at app.godriftless.ai, the Driftless REST API, and the Driftless MCP endpoint (collectively, the “Service”) for all customers of the hosted Service.
We commit to maintaining a Service uptime of 99.5% per calendar month, excluding the downtime events described in the Exclusions section (“Downtime”). Measured uptime has been 100% since external monitoring went live in June 2026; we publish 99.5% because a target is only meaningful if we can keep it.
This SLA applies only to the hosted Service. It does not apply to any beta, preview, or pre-release features, to the marketing website at godriftless.ai, or to any functionality explicitly marked as not covered by this SLA.
How uptime is measured
Uptime is measured over a rolling calendar-month window by external synthetic monitoring (Better Stack) polling the Service’s public health endpoints at 30-second intervals, around the clock.
What counts as Downtime
“Downtime” means a continuous period of one minute or longer during which the external monitoring could not obtain a successful response from a public health endpoint of the Service, where the unavailability is not caused by an excluded event under Exclusions.
- Measurement: Downtime is recorded from the time monitoring detects the unavailability until the Service responds successfully again.
- Window: the measurement window is the calendar month in UTC. Minutes during which an excluded event was in effect are subtracted from both Downtime and the total minutes in the month.
- Evidence: monitoring results are published live on the public status page, including incident history.
Uptime calculation
Monthly uptime percentage is calculated as:
Support response times
We classify support incidents by severity. The severity matrix below defines each level and the target response time, measured from when a valid incident report is received until a support engineer begins working on it.
| Severity | Definition | Response time |
|---|---|---|
| P1: Critical | Service unavailable: the Service is down or core functionality is inaccessible to all or most users. | 1 hour |
| P2: High | Major feature degraded: a core feature is significantly impaired but a workaround exists or impact is partial. | 4 hours |
| P3: Medium | Minor feature issue: a non-core feature is impaired or usability is reduced, with a reasonable workaround available. | 1 business day |
| P4: Low | Cosmetic or non-urgent: cosmetic defects, documentation requests, or feature requests with no operational impact. | 2 business days |
How to report an incident
To report a support incident, contact us at support@godriftless.ai with your organization name, the affected component, a description of the issue, and its business impact.
Incident notification
When we become aware of an incident that affects Service availability, we will notify affected customers without undue delay. Notification includes the nature of the incident, the affected components, and an estimate of impact.
- Initial notification: within 30 minutes of incident confirmation for P1 and P2 incidents, and as soon as practical for P3 and P4 incidents.
- Updates: we will provide status updates at least every 2 hours during an active P1 or P2 incident until the incident is resolved.
- Resolution notice: a summary of the incident, root cause (to the extent known), and corrective actions is provided after resolution.
Status page
Real-time Service status is published at status.godriftless.ai, including current status, incident history, and maintenance notices. Customers are encouraged to subscribe to status updates there.
Communication channels
Incident notifications are sent via email to account owners and published on the status page. Customers are responsible for keeping their contact information current so that we can reach them during incidents.
Exclusions
The uptime commitment and Downtime calculations do not include unavailability caused by the following events. Downtime during these events does not count against the uptime target:
- Scheduled maintenance. Downtime during pre-announced scheduled maintenance windows. We will provide at least 72 hours notice of scheduled maintenance and will, where practicable, schedule it during off-peak hours.
- Customer-caused issues. Issues caused by the customer’s or its users’ actions or inactions, including misconfiguration, incorrect credentials, exceeded quotas, custom integrations, unsupported client software, or failure to follow documented guidance.
- Third-party outages. Unavailability of third-party services or dependencies outside Driftless’s control, including cloud-provider regional outages, DNS providers, identity providers, and customer-supplied API providers, to the extent the Service depends on them.
- Force majeure. Events beyond our reasonable control, including natural disasters, acts of war or terrorism, labor disputes, governmental actions, and widespread internet or power outages.
- Factors outside our infrastructure. Failures of the internet backbone, customer-side network or equipment failures, or issues with the customer’s ability to reach the Service over the public internet.
Changes to this SLA
We may update this SLA from time to time. When we do, we will revise the “Last updated” date above. If we make material changes that reduce the uptime commitment or otherwise weaken the commitments in this SLA, we will provide commercially reasonable notice, for example by email or an in-app notice, before the changes take effect.
Contact
Questions about this SLA can be directed to support@godriftless.ai.