Service Level Agreement

Effective Date: January 16th, 2026

This Service Level Agreement (“SLA”) forms part of and is incorporated into the Deyra Terms of Service or the Master Service Agreement between Five Jars Corp (“Five Jars”, “we”, “us”, “our”), the provider of the Deyra service (“Deyra”), and the Customer (“you”, “your”). Deyra is a product of Five Jars Corp. Capitalized terms not defined here have the meanings given in the applicable agreement.

1. Scope

This SLA applies to the production Deyra AI Agent and data orchestration services (including the AI Agent interface, embeddable search widget (Data Finder), Web UI, APIs, data integration services, and query processing capabilities) (collectively, the “Service”). This SLA does not apply to:

• Features or services marked alpha/beta/preview, or otherwise unsupported;
• Free plans or trials unless explicitly stated;
• Third‑party networks, ISPs, content delivery networks (CDNs) outside our reasonable control;
• Third-party data sources, CRM systems, databases, or APIs that you integrate with Deyra;
• Customer-side network connectivity, firewalls, or authentication systems;
• Data accuracy or completeness issues originating from your source systems.

2. Definitions

• “Monthly Uptime Percentage” (MUP) means: MUP = (Total minutes in month − Downtime minutes) ÷ (Total minutes in month) × 100%
• “Downtime” means periods when the AI Agent or core Service APIs are unavailable to process end-user queries or data requests from the public internet due to a Deyra‑controlled incident, measured in continuous 5‑minute intervals. Partial intervals under 5 minutes are not counted as Downtime.
• “Business Hours” means Monday–Friday, 09:00–18:00 Eastern Time (excluding U.S. federal holidays), unless otherwise agreed.
• “Planned Maintenance” means maintenance we schedule with at least 72 hours’ notice.
• “Emergency Maintenance” means urgent maintenance required to address active incidents or security issues.
• “AI Agent” means the natural language interface that processes user queries and returns results from connected data sources.
• “Data Integration Services” means the connectivity, data synchronization, and orchestration capabilities that connect your data sources to the AI Agent.

3. Service Availability Commitment

We will use commercially reasonable efforts to provide a Monthly Uptime Percentage of at least 99.95% for the Service.

3.1 Measurement

• Availability is measured at the Service edge (Deyra‑controlled ingress) using our monitoring systems for both AI Agent query processing and API endpoints.
• Downtime excludes: Planned Maintenance, Emergency Maintenance (limited and reasonable), force majeure events, internet/general connectivity issues outside our control, third‑party data source failures (CRM systems, databases, external APIs), Customer misuse or misconfiguration, integration authentication failures on your systems, and violations of the AUP/Terms.

3.2 Maintenance Windows

• Planned Maintenance is typically scheduled Sundays 02:00–06:00 UTC and will be announced at least 72 hours in advance.
• We strive to perform zero‑downtime maintenance; where service impact is anticipated, we will provide expected impact and duration.

4. Incident Response Targets

Five Jars provides support for the Deyra service during Business Hours, with 24×7 response for Severity 1 incidents.

Severity	Example	Initial Response	Status Updates	Target Workaround/Resolution
Sev 1 – Critical	Complete AI Agent outage; Core APIs unavailable; All data integrations failing; No query responses	1 hour (24×7)	Hourly until mitigated	Workaround target: 4 hours; Resolution: commercially reasonable efforts
Sev 2 – High	Significant degradation; Multiple integrations failing; Slow query response times; Regional impact	4 business hours	Every business day	Workaround/Resolution: commercially reasonable efforts
Sev 3 – Medium	Single integration issues; Limited feature impairment; Non‑critical bugs; UI issues	1 business day	Weekly	Added to backlog; prioritized per impact
Sev 4 – Low	Cosmetic issues; Documentation requests; Feature questions	2 business days	As needed	Handled via standard support process

Note: Targets are response commitments, not guarantees of resolution time.

5. Service Credits

If the Monthly Uptime Percentage falls below the commitment, you may be eligible for a service credit, calculated against the monthly subscription fees for the affected Service and region.

Monthly Uptime Percentage (MUP)	Credit
< 99.95% and ≥ 99.9%	5%
< 99.9% and ≥ 99.0%	10%
< 99.0% and ≥ 98.0%	25%
< 98.0%	50%

Credit Terms:

• Credits apply to future invoices and cannot be refunded in cash.
• Credits are capped at 50% of the monthly fee for the affected month and Service.
• No credits for free plans/trials or for periods when account payments are overdue.
• Credits are the sole and exclusive remedy for failure to meet this SLA.

5.1 Requesting Credits

To receive a credit, you must submit a request to [email protected] within 30 days after the month in which the SLA was not met. The request must include:

• Account name and billing email;
• Dates, times, and regions of the suspected Downtime;
• A description of the business impact;
• Any relevant logs or incident IDs.

We will validate using our logs and monitoring data.

6. Exclusions

Downtime does not include unavailability, suspension, or termination of the Service caused by:

• Factors outside our reasonable control (force majeure, DDoS, large‑scale internet outages);
• Failures of Customer network connectivity, firewalls, VPNs, ISPs, or third‑party services/CDNs;
• Third-party data source unavailability, including CRM systems, databases, or external APIs you integrate with Deyra;
• Authentication or authorization failures with your integrated systems;
• Data quality, accuracy, or completeness issues originating from your source systems;
• Use of beta/preview features or non‑GA functionality;
• Actions or inactions by Customer or users in violation of the Agreement or AUP;
• Rate limiting or resource exhaustion due to Customer configuration or traffic spikes beyond contracted capacity without prior notice;
• Maintenance performed per Section 3.2.

7. Security and Data Protection

We maintain administrative, physical, and technical safeguards designed to protect the Service and Customer Data. Additional details are available in our Privacy Policy and Security documentation. Security incidents are handled under our incident response procedures. For data processing, our DPA applies where executed.

• Privacy Policy
• Terms of Service

8. Customer Responsibilities

• Maintain accurate contact information and designate incident contacts;
• Ensure your integrated data sources (CRMs, databases, APIs) maintain stable connectivity and authentication;
• Provide accurate API credentials and access permissions for data integrations;
• Maintain data quality and accuracy in your source systems;
• Configure the Service according to documentation and best practices;
• Use the Service within contracted capacity and notify us of material forecasted increases in query volume or data sources.

9. Changes to this SLA

We may update this SLA from time to time. Material changes will be communicated via our status page, dashboard notice, or email to your admin contacts at least 30 days in advance, unless changes are required for security or compliance and reasonably cannot be delayed.

10. Scope Limitation - AI Response Quality

What This SLA Covers:

This SLA applies ONLY to:

• Service availability (uptime/downtime of the platform)
• API response times and processing capabilities
• Data integration connectivity and synchronization
• Platform accessibility from the public internet

What This SLA Does NOT Cover:

This SLA does not cover, and no service credits, refunds, or other remedies are provided for:

AI Output Quality:

• Accuracy, completeness, relevance, or appropriateness of AI-generated responses
• Quality, correctness, suitability, or reliability of AI-generated content
• AI hallucinations, fabrications, confabulations, or false information
• AI misinterpretations, incorrect reasoning, or flawed synthesis of data
• AI biases, discriminatory outputs, or inappropriate content
• Relevance or ranking of search results in Data Finder
• Variability or inconsistency in AI responses to similar queries

Data Interpretation:

• How the AI Agent interprets, understands, or presents data from your connected sources
• AI-generated summaries, analyses, or insights derived from your data
• Conclusions, recommendations, or suggestions made by the AI
• Contextual appropriateness of AI responses for specific user queries

Business Outcomes:

• Business decisions, strategies, or actions based on AI outputs
• Revenue losses, missed opportunities, or operational inefficiencies related to AI response quality
• Customer satisfaction, retention, or conversion impacts from AI interactions
• Reputational harm or brand damage from AI-generated content visible to your customers

Regulatory and Compliance:

• Your compliance with AI disclosure laws or regulations
• Claims by regulatory authorities related to your AI deployment
• Violations of professional standards or licensing requirements
• Consumer protection issues arising from AI-generated content

Third-Party Claims:

• Claims by your customers or end-users regarding AI response quality
• Disputes arising from actions taken based on AI-generated information
• Allegations of harm from inaccurate or inappropriate AI outputs

Exclusive Remedy:

All matters related to AI response quality, accuracy, appropriateness, and limitations are governed exclusively by:

• The “AI Services and Limitations” section of the Terms & Conditions
• The “AI Output Disclaimer and Non-Reliance” section of the Terms & Conditions
• The “Disclaimer of Warranties” section of the Terms & Conditions
• The “Limitation of Liability” section of the Terms & Conditions

No Monitoring Obligation:

Five Jars does not monitor and is not responsible for:

• The quality, accuracy, or appropriateness of individual AI responses
• How customers deploy or present AI-generated content to their end-users
• Customer verification, review, or validation procedures (or lack thereof)
• Customer compliance with implementation requirements or disclosure obligations

Customers are solely responsible for:

• Reviewing and validating AI outputs before relying on them or presenting them to end-users
• Implementing appropriate monitoring, logging, and quality control procedures
• Ensuring AI deployments are appropriate for their specific use cases and risk tolerance
• Complying with all applicable laws regarding AI disclosure, consumer protection, and professional standards

You acknowledge that AI quality issues are not service availability issues and are not covered by this SLA. Any attempt to claim service credits for AI quality issues will be denied.

11. Relationship to Agreement; Sole Remedy

This SLA is part of the Agreement and is subject to its limitations and exclusions. THE SERVICE CREDITS PROVIDED IN SECTION 5 ARE YOUR SOLE AND EXCLUSIVE REMEDY FOR ANY FAILURE TO MEET THE UPTIME COMMITMENTS IN THIS SLA. TO THE MAXIMUM EXTENT PERMITTED BY LAW, FIVE JARS CORP DISCLAIMS ALL OTHER WARRANTIES, REMEDIES, AND LIABILITIES RELATED TO SERVICE AVAILABILITY. You waive any right to seek damages, refunds, or other remedies beyond the service credits specified herein for SLA failures.

12. Contact and Support

• Support: [email protected]
• Status page: https://status.deyra.ai
• Emergency escalations (Sev 1): Provided after ticket creation if requested

13. Customer Monitoring and Incident Reporting

Your Monitoring Obligations:

When deploying AI Agent or Data Finder on customer-facing properties, you are responsible for:

a) Ongoing Monitoring:

• Implementing logging and review procedures appropriate to your use case
• Regularly sampling AI responses visible to your customers
• Monitoring for patterns of errors, inappropriate outputs, or quality degradation
• Tracking customer complaints or concerns about AI response quality

b) Incident Identification:

You must identify and report to Five Jars within 48 hours any incidents involving:

• Systematic errors or repeated inaccuracies across multiple queries
• Discriminatory, offensive, or harmful outputs
• Disclosure of sensitive information or potential privacy violations
• Outputs that could create legal liability or regulatory exposure
• Consistent failures to retrieve or interpret data from connected sources
• Patterns suggesting prompt injection, adversarial attacks, or abuse

c) Incident Reporting:

Reports must include:

• Description of the issue and observed pattern
• Example queries and AI responses demonstrating the problem
• Approximate frequency and duration of the issue
• Impact on your operations or customers
• Any relevant logs, screenshots, or conversation IDs
• Steps you have taken to mitigate customer impact

Five Jars’ Response:

Upon receiving a valid incident report, Five Jars will:

• Acknowledge receipt within 1 business day
• Investigate the reported issue using our monitoring data and logs
• Determine if the issue represents a platform-level problem or is specific to your deployment configuration
• Provide findings and recommendations within a commercially reasonable timeframe

Five Jars’ obligation is to investigate and respond, not to guarantee resolution of AI quality issues. Many reported “issues” may be inherent limitations of AI technology rather than service defects.

Not Service Availability Issues:

You acknowledge that:

• AI quality issues reported under this section are not service availability failures under Section 3
• No service credits are available for AI quality issues
• Five Jars’ investigation and response is provided as a courtesy to help customers understand and manage AI limitations
• Five Jars retains sole discretion to determine whether an issue warrants platform-level changes

Failure to Monitor:

Your failure to implement appropriate monitoring and reporting procedures:

• Does not reduce your responsibilities under the Terms & Conditions
• Does not extend the time for reporting incidents or making claims
• May be considered in determining liability for customer-facing AI issues
• Does not excuse your indemnification obligations for third-party claims

---