API & SDK Usage Policy

This API and SDK Usage Policy governs the use of Cortexvia's application programming interfaces (APIs) and software development kits (SDKs). Cortexvia is operated by Shaik Afrid, trading as Cortexvia Technologies ("Cortexvia," "we," "us," or "our"). Cortexvia provides programmatic access to our AI chatbot platform, enabling developers to integrate advanced natural language processing capabilities into their applications, websites, and services. By accessing our APIs or using our SDKs, you agree to comply with this Policy and all applicable terms within the Agreement. This policy should be read in conjunction with our Terms of Service, Privacy Policy, Data Usage Policy, Security Policy, Cookie Policy, Compliance Policy, Developer Policy, and Refund Policy (collectively, the "Agreement").

1. API Key Management and Security

1.1. Key Issuance. Upon account creation and subscription to an appropriate tier, users are issued API keys that authenticate requests to the Cortexvia API. API keys are unique to each account and should be treated as sensitive credentials. 1.2. Multi-Site Usage. Users may use the same API key across multiple websites, applications, or integrations. However, all usage across all sites counts toward the single quota associated with that API key. 1.3. Secure Storage. Cortexvia stores API keys securely using encryption and access controls. However, once an API key is copied or displayed in the dashboard, it is entirely the user's responsibility to maintain its confidentiality and security. 1.4. Post-Issuance Responsibility. We are not responsible for any unauthorized usage, security breaches, quota exhaustion, or other issues arising from compromised API keys after they leave our dashboard. Users bear full responsibility for the security of their credentials. 1.5. Storage Best Practices. Never share API keys publicly, in forums, or in public code repositories. Do not commit API keys to version control systems such as Git. Store keys in environment variables or secure secret management systems, never in application source code. Never expose API keys in frontend, client-side, or browser-accessible code where they can be extracted. 1.6. Key Rotation. Users should regularly rotate API keys as a security best practice. Keys suspected of being compromised should be immediately revoked and replaced through the dashboard.

2. Rate Limits and Usage Quotas

2.1. Tier-Based Limits. Rate limits and usage quotas are determined by your subscription tier. These limits are described on our pricing page and may be adjusted at our discretion with appropriate notice. 2.2. Quota Consumption. Using one API key across multiple sites or applications causes quota to be consumed faster. Users are responsible for monitoring their usage and staying within their allocated limits. 2.3. Rate Limit Responses. Requests that exceed rate limits receive HTTP 429 (Too Many Requests) responses. These responses include headers indicating when the rate limit resets and how many requests remain. 2.4. Backoff Requirements. Clients must implement exponential backoff when receiving rate limit errors. Aggressive retry patterns that ignore rate limit responses may result in temporary or permanent API access suspension. 2.5. Limit Increases. If you consistently require higher limits than available in your current tier, contact support to discuss upgrade options or enterprise arrangements with custom quotas.

3. Prohibited Uses and Content Restrictions

3.1. Absolute Prohibitions. The following uses are absolutely prohibited and will result in immediate account suspension without warning: integration with gambling websites or applications, integration with adult content or pornographic websites, integration with websites promoting hate speech, violence, or discrimination, and any use that violates applicable laws. 3.2. Harmful Content. Users shall not use the API to create chatbots that generate illegal, harmful, threatening, abusive, harassing, defamatory, or otherwise objectionable content. 3.3. Impersonation. Building chatbots that impersonate real people, organizations, or government entities without explicit authorization is prohibited. 3.4. Spam and Abuse. Using APIs for spamming, phishing, unsolicited communications, or any form of automated abuse is prohibited. 3.5. Reverse Engineering. Attempting to reverse engineer, decompile, disassemble, or otherwise derive source code or algorithms from our API responses or systems is prohibited. 3.6. Security Circumvention. Bypassing, circumventing, or attempting to circumvent rate limits, authentication mechanisms, or other security measures is prohibited. 3.7. Resale Prohibition. Selling, sublicensing, or otherwise commercially redistributing API access to third parties without explicit written authorization is prohibited. 3.8. System Abuse. Overloading our systems with excessive, unnecessary, or malicious requests that degrade service quality for other users is prohibited.

4. Account Suspension and Termination

4.1. Suspension Phase. Upon detection of policy violations, accounts may be immediately suspended without prior notice. During suspension, all API keys associated with the account stop functioning. Dashboard access may be restricted, though billing modules may remain active to allow settlement of outstanding charges. 4.2. Cure Period. Suspended users have thirty (30) days from the date of suspension notification to provide an explanation, resolve the violation, or appeal the decision. Communication regarding suspension should be directed to [email protected]. 4.3. Permanent Deletion. If no satisfactory resolution is provided within the thirty-day cure period, the account and all associated data will be permanently deleted without further notice or opportunity for recovery. 4.4. No Refunds. No refunds will be issued for subscription fees, unused quota, or any other charges associated with accounts suspended or terminated due to policy violations. Any remaining subscription time is forfeited upon termination for cause. 4.5. Legal Action. Egregious violations may be referred to law enforcement authorities and may result in civil or criminal legal action in addition to account termination.

5. Third-Party Content Sources

5.1. YouTube Content. The Services enable extraction and storage of text and transcripts from YouTube videos for use in chatbot response generation and retrieval. When content is extracted, only the text transcript is stored; video files are not downloaded or retained. 5.2. Source Availability. If a YouTube video is subsequently deleted, made private, or blocked by the content owner, the stored transcript remains available but chatbot responses may reference broken or unavailable source links. This is not the responsibility of Cortexvia. 5.3. User Rights Responsibility. Users are solely responsible for ensuring they have the legal right to use content from third-party sources, including YouTube videos, for chatbot response generation purposes. Users should consult the terms of service of the original content sources. 5.4. Source Management. Users can enable, disable, or remove specific content sources for their chatbots at any time. If a source becomes unavailable, users should update or remove it to maintain response quality. 5.5. Link Accuracy. While we attempt to maintain accurate source references, we cannot guarantee the continued availability or accuracy of links to external content. Chatbot responses referencing external sources should be verified by end-users when accuracy is critical.

6. Inference Pipeline and Response Generation

6.1. Pipeline Architecture. API requests for AI-generated responses trigger a multi-stage inference pipeline that includes query processing, context retrieval, prompt construction, and response generation. Each stage involves computational operations that may be performed by different components of our distributed system. 6.2. Retrieval-Augmented Generation. The Services use retrieval-augmented generation (RAG) techniques that combine information retrieval from your uploaded content with large language model inference to generate contextually relevant responses. 6.3. Computational Infrastructure. Response generation involves transmission of query context to specialized computational endpoints optimized for large language model inference. These endpoints may utilize external infrastructure providers operating GPU-accelerated computing clusters. The specific infrastructure providers and models utilized may change without notice as we optimize for performance, cost, and capability. 6.4. Model Provider Variations. Different API requests may be processed by different underlying model providers depending on load balancing, model availability, and optimization algorithms. This may result in subtle variations in response style or capability across requests. 6.5. Processing Acknowledgment. By making API requests, you acknowledge that request data including the query and retrieved context is processed by our inference pipeline, which may involve transmission to and processing by external computational service providers operating under their own data handling policies and procedures. For production workloads, provider configurations are intended for inference-only processing, while provider retention and model-improvement terms depend on provider contracts and service tiers.

7. Chatbot End-User Data and Liability

7.1. Conversation Storage. Questions and answers from chatbot conversations are stored and associated with the chatbot owner's account. This data is accessible through the dashboard and via API endpoints. 7.2. End-User Data Responsibility. If end-users (website visitors interacting with your chatbot) enter sensitive information in chat conversations, the chatbot owner is fully responsible for handling this data appropriately. Cortexvia acts as a data processor for this end-user data; the chatbot owner is the data controller. 7.3. Owner Liability. The chatbot owner is fully responsible and liable for their chatbot's questions, answers, and any consequences arising from chatbot interactions. This includes responsibility for legal compliance, accuracy of responses, and any claims brought by end-users. 7.4. Data Retention Control. Chatbot owners control how long conversation data is retained and can delete conversation history at any time through the dashboard or API. 7.5. AI Response Limitations. AI models can produce inaccurate, incomplete, biased, or unexpected responses. Cortexvia does not guarantee the accuracy, reliability, or appropriateness of any AI-generated content. Users should implement appropriate safeguards and disclaimers for their end-users. 7.6. Liability Disclaimer. Cortexvia is not responsible or liable for any harm, loss, damage, claims, or legal issues caused by chatbot responses, including but not limited to misinformation, incorrect advice, offensive content, or business decisions made based on AI-generated content.

8. Beta Features and Service Modifications

8.1. Experimental Features. Beta, experimental, or preview features may be offered for testing purposes. These features are provided "as is" without warranty or support commitment and are not recommended for production use. 8.2. Beta Removal. Beta features can be removed, modified, or discontinued at any time without prior notice, email notification, or transition period. Applications dependent on beta features may break without warning. 8.3. Production Feature Changes. For stable, production-grade features, we endeavor to provide minimum thirty (30) days notice before removal or major breaking changes. Notice will be provided through email, dashboard announcements, or API deprecation headers. 8.4. General Modification Rights. We reserve the right to modify, update, enhance, or discontinue any aspect of the API at any time. Major changes are not made frequently and only after careful consideration, but operational requirements may necessitate changes.

9. Infrastructure Scalability

9.1. Horizontal Scaling. The API infrastructure is designed for horizontal scalability. We may add servers, computational nodes, and inference capacity without notice to handle growth and traffic spikes. 9.2. Modular Architecture. Services are architected to support modular expansion, allowing individual components to scale independently. This ensures that increased load on one component does not degrade performance of others. 9.3. Redundancy Limitations. While we maintain redundancy at critical layers, we do not guarantee zero-downtime or perfect availability. Planned maintenance and unexpected issues may cause temporary service interruptions. 9.4. Evolution Notice. Infrastructure architecture may evolve over time to improve performance, security, and reliability. Such evolution may occur without prior notice to users.

10. Support Policy

10.1. Email Support. Email support is available for all users at [email protected]. All support requests will receive an initial response within twenty-four (24) hours during business days. 10.2. Phone Support. Phone support is available only for Startup, Business, and Student subscription tiers. Phone support is provided in English, and our team can assist in most major languages on request. 10.3. Resolution Timeline. We target final resolution of support issues within seven (7) business days, though complex technical issues may require additional time. 10.4. Support Abuse. Misuse of support channels, including harassment of support staff, excessive frivolous requests, or use of support phone numbers for non-support purposes, may result in support access restrictions or account suspension.

11. Contact Information

If you have any questions about this API Policy, please contact us at:

Cortexvia
General Support: [email protected]
Legal Inquiries: [email protected]
Billing Issues: [email protected]

By using the Cortexvia API, you acknowledge that you have read and understood this API Policy and agree to be bound by its terms.