CloudFran Agents Platform Addendum

This Agents Platform Addendum ("Addendum") supplements the CloudFran Terms of Service and governs the use of all AI agents, automation tools, and workflow capabilities available on the agents.cloudfran.com platform ("Agents Platform"). This Addendum is incorporated by reference into the Master Services Agreement.

Individual agent products (Autopilot, VoicePersuader, OmniReceptionist, etc.) have their own product-specific addenda that supplement this platform-level document. In the event of conflict between this Addendum and a product-specific addendum, the product-specific addendum controls for that product.

1. Scope & Platform Definition

The Agents Platform is a unified AI-powered environment hosted at agents.cloudfran.com that provides access to multiple AI agents as paid upgrade services. Each agent is a distinct capability module that can operate autonomously or semi-autonomously within user-configured parameters.

The Agents Platform includes, but is not limited to:

Business Advisor — strategic guidance and franchise economics analysis
Content Writer — AI-generated marketing copy, social posts, emails, and menus
Data Analyzer — spreadsheet analysis with AI-driven insights and visualizations
Image Generator — branded marketing graphics and social media assets
Employee Coaching — AI-generated manager coaching playbooks
Campaign Autopilot — AI-driven outbound campaigns (see Autopilot Addendum)
Voice Persuader — IVR tree design and voice workflows (see VoicePersuader Addendum)
OmniReceptionist — AI voice receptionist (see OmniReceptionist Addendum)

CloudFran may add, modify, or discontinue individual agents at any time. The availability of specific agents may vary by subscription tier.

2. Agent Types & Capabilities

Agents on the platform fall into three operational categories:

Advisory agents — provide analysis, recommendations, or generated content that requires user review before action (Business Advisor, Data Analyzer, Content Writer)
Action agents — execute automated actions on behalf of the user within configured parameters (Campaign Autopilot, Employee Coaching)
Communication agents — interact with external parties (callers, email recipients, SMS recipients) on behalf of the user (OmniReceptionist, VoicePersuader, Autopilot outbound)

Communication agents carry the highest compliance obligations because they interact with third parties. You are the principal and the agent acts on your behalf — you bear responsibility for the agent's communications as if you had sent them yourself.

3. AI Output Disclaimer

AI outputs are probabilistic, not deterministic. All agent outputs — including text, analysis, recommendations, voice responses, images, and decisions — may contain errors, omissions, inaccuracies, or biases. CloudFran does not guarantee the accuracy, completeness, legality, or appropriateness of any AI output.

You acknowledge that:

AI models may produce different outputs for identical inputs
AI recommendations are not professional advice (legal, financial, medical, tax, or otherwise)
AI-generated content may inadvertently infringe third-party intellectual property rights
AI models are updated periodically, which may change behavior and outputs

4. Human Validation Requirement

For advisory and action agents, you are responsible for reviewing AI outputs before relying on them for business decisions, publishing them to customers, or allowing them to trigger downstream actions.

For communication agents, you are responsible for configuring the agent's knowledge base, scripts, escalation rules, and guardrails. Once configured, the agent operates autonomously within those parameters — you bear responsibility for the agent's behavior within your configured scope.

CloudFran strongly recommends enabling human-in-the-loop review for high-stakes agent actions (financial decisions, legal communications, healthcare interactions, employment decisions).

5. Workflow Responsibility

You are solely responsible for:

Designing and configuring agent workflows
Setting appropriate guardrails, limits, and escalation triggers
Monitoring agent outputs and correcting errors
Ensuring workflows comply with applicable laws and regulations
Testing workflows before deploying them to production

CloudFran provides the platform, tools, and AI capabilities. You design the workflow and bear responsibility for its outcomes.

6. Messaging & Communications Compliance

If any agent sends email, SMS, push notifications, or other electronic communications on your behalf, you are responsible for compliance with:

CAN-SPAM Act (United States) — accurate headers, unsubscribe link, physical address, honor opt-outs within 10 business days
TCPA (United States) — prior express written consent for marketing SMS/calls to mobile numbers
CASL (Canada) — express opt-in consent before commercial electronic messages
GDPR / ePrivacy (EU/UK) — lawful basis for processing, consent or legitimate interest
Equivalent laws in all other jurisdictions where your recipients reside

CloudFran provides compliance tools (unsubscribe links, suppression lists, consent tracking) but does not guarantee your compliance. You are the sender.

7. Voice & Telephony Agents

Agents that place or receive phone calls (OmniReceptionist, VoicePersuader) rely on third-party telephony providers including Twilio. You acknowledge:

Call quality, routing, and delivery depend on carriers and are not controlled by CloudFran
Call recording requires compliance with state-by-state consent laws (all-party consent in CA, FL, IL, MA, MD, MT, NH, PA, WA; one-party consent in most other states)
AI voice disclosure may be required by law (e.g., FTC guidelines on AI-generated voice, state AI disclosure laws)
Do Not Call (DNC) list compliance is your responsibility for outbound campaigns

See the OmniReceptionist Addendum and VoicePersuader Addendum for product-specific voice terms.

8. Cross-Product Integration

Agents may integrate with other CloudFran products via shared data layers (e.g., CrmAutopilotBridge connects Campaign Autopilot with CloudFran CRM). When cross-product integration is active:

Data flows between products within your tenant's isolation boundary
CloudFran does not share agent data across tenants
You may disable cross-product integrations in your settings
Each product's own addendum governs its specific terms — the integration does not merge the terms

Cross-product data sharing is an optional feature. Tenants with only one product see no data from products they don't subscribe to.

9. Fees & Billing

Each agent is billed according to its subscription tier as displayed on the agent's pricing page. In addition:

Transactions processed through agent workflows are subject to CloudFran platform fees of 3.5% + $0.75 per transaction where applicable
AI token consumption (for agents that use large-language-model calls) may be metered and billed per your plan's included allocation, with overage rates disclosed in your billing dashboard
Telephony minutes, SMS segments, and carrier fees are passed through at cost plus applicable markup
All fees are non-refundable except where required by law

10. Chargebacks & Disputes

CloudFran mirrors payment processor protections, including dispute fees of $15 where applicable. You are responsible for chargebacks, disputes, and all associated losses arising from transactions initiated or facilitated by agents acting on your behalf.

11. Data Usage & AI Training

You acknowledge that interaction data (inputs, outputs, configurations, usage patterns) may be processed for:

Service delivery and agent operation
Quality improvement and error correction
AI model training and optimization (in aggregated, anonymized form)
Analytics and reporting within your tenant dashboard
Fraud detection and platform security

Data processing is governed by the Privacy Policy and Data Processing Agreement. If you enable HIPAA mode for healthcare agents, PHI handling is governed by your BAA.

12. Acceptable Use & Prohibited Activities

You may not use any agent on the Agents Platform to:

Generate spam, phishing, or deceptive content
Impersonate individuals, organizations, or government entities
Create content that promotes violence, discrimination, or illegal activity
Circumvent AI safety guardrails or content filters
Process data in violation of applicable privacy laws
Conduct automated interactions that violate TCPA, CAN-SPAM, or DNC rules
Use AI outputs as professional advice (legal, medical, financial, tax) without qualified human review
Reverse-engineer, extract, or replicate CloudFran's AI models or training data

CloudFran may suspend or terminate agent access for violations, with notice as soon as practicable.

13. International Restrictions

Certain agent capabilities may be restricted by geography due to regulatory requirements, AI governance laws, or third-party provider limitations. Restrictions include but are not limited to:

EU AI Act high-risk classification for certain agent types (employment, credit, healthcare)
Country-specific data residency requirements
Telephony regulations that vary by carrier and jurisdiction
Payment processing availability per market

CloudFran may modify agent availability per jurisdiction without liability.

14. Liability & Indemnification

To the maximum extent permitted by law, CloudFran is not liable for:

AI output errors, omissions, or inaccuracies
Business decisions made based on agent recommendations
Communications sent by agents on your behalf
Regulatory violations arising from your agent configurations
Lost revenue, missed opportunities, or reputational harm
Third-party service failures (Twilio, Stripe, AI model providers)

Liability cap: CloudFran's total aggregate liability under this Addendum shall not exceed the fees you paid for agent subscriptions in the 3 months preceding the claim.

Indemnification: You agree to indemnify and hold harmless CloudFran from claims arising from your agent configurations, workflows, communications, AI output usage, and regulatory non-compliance.

15. Governing Law & Compliance

This Addendum is governed by Delaware law with Georgia enforceability considerations. All disputes shall be resolved via binding arbitration (AAA) with a class action waiver, as set forth in the Master Services Agreement.

CloudFran aligns with SOC2 security controls, AI governance best practices, communications compliance standards, and fintech audit readiness.

16. Changes to This Addendum

CloudFran may update this Addendum from time to time. Material changes will be communicated at least 30 days in advance via email or dashboard notification. Continued use of the Agents Platform after the effective date of any update constitutes acceptance.