Acceptable Use Policy

Do not use the Service to plan, facilitate or carry out unlawful activity, or to store or generate content that is unlawful where you or we operate. This includes trafficking in controlled substances, weapons or stolen data, money laundering, evading sanctions, and unlicensed regulated activity.

Do not use the Service to harass, bully, stalk, threaten or intimidate anyone, to incite violence, to dox or reveal private information about a person, or to generate content that demeans or dehumanises people on the basis of a protected characteristic.

Do not use the Service to promote, glorify, coordinate or provide material support for terrorism or violent extremism, to recruit for a violent organisation, or to plan attacks on people, property or infrastructure. Do not use it to design weapons, explosives, or chemical, biological, radiological or nuclear capability.

Do not use the Service to write, refine, obfuscate, test or distribute malware, ransomware, spyware, keyloggers, botnets, cryptominers deployed without consent, or exploit code intended for unauthorised access. Legitimate security work on systems you own or are authorised to test is permitted; see Section 4.4.

Do not use the Service to build or draft phishing pages, credential-harvesting flows, fake storefronts, deceptive payment pages, scam scripts, or fraudulent financial schemes. Do not impersonate another person or organisation, misrepresent your affiliation, or use another person's identity documents or personal data without authorisation.

Do not use the Service to generate or manage bulk unsolicited messaging, to build systems whose purpose is evading spam filters or consent requirements, or to produce mass low-quality content intended to manipulate search rankings or platform recommendations.

Do not upload, generate or distribute material that infringes copyright, trademark, patent or trade secret rights, and do not use the Service to remove watermarks, strip attribution, circumvent technical protection measures, or launder proprietary code or content whose licence you cannot comply with.

Do not upload personal data you have no lawful basis to process, scraped datasets assembled in breach of another platform's terms, biometric identifiers, or special-category data. Do not use the Service to build tools for covert surveillance, unlawful tracking, or facial recognition of people without consent.

Do not reverse engineer, decompile or disassemble the Service, attempt to extract our source code, system prompts, manifests, planning logic or model configuration, or copy the Service to create a substantially similar product. This restriction does not apply where it is prohibited by applicable law.

Do not attempt to access an Account, Project, dataset, API or system you are not authorised to access; probe or bypass authentication or authorisation controls; interfere with another user's use of the Service; or interfere with our logging and monitoring.

Do not place a disproportionate load on the Service, run stress or load tests against it, open excessive concurrent sessions, or participate in a denial-of-service attack. Do not use the Service to run workloads unrelated to specifying and planning your own product, for example general-purpose bulk text generation.

Do not conduct penetration testing, vulnerability scanning, fuzzing or automated exploitation against Zagvo or our providers without our prior written permission.

We welcome good-faith security research. Report suspected vulnerabilities privately to legal@zagvo.com, give us reasonable time to fix before disclosure, do not access or exfiltrate other users' data, and do not degrade the Service. Research conducted on those terms will not be treated as a Policy violation.

Do not use bots, scrapers, headless browsers or scripted clients to interact with the Service other than through interfaces we document for that purpose. Do not create Accounts automatically, evade bot protection, or solve or bypass challenges programmatically.

Accounts are for one person. Do not share credentials, sell, rent or transfer an Account, use another person's Account, or operate multiple Accounts to obtain additional Credits, extend a trial, or evade an enforcement action.

Where we make an interface available, use it within the documented limits, identify your client honestly, cache responsibly, and back off when asked to. Do not attempt to raise your effective throughput by rotating Accounts, IP addresses or keys, and do not resell access to our AI capability.

Do not tamper with Credit metering, replay or forge requests to obtain uncharged AI usage, manipulate subscription state, or abuse refunds, promotional codes or beta access to obtain service you have not paid for.

Do not use the Service to build a competing product, to systematically extract our prompts, manifests or planning behaviour for replication, or to publish benchmarks or comparisons derived from such extraction without our written consent.

Do not attempt to bypass, disable or trick the safety measures of the Service or of the models behind it, including jailbreaking, encoding prohibited requests to evade filters, using role-play to elicit prohibited output, or chaining requests so that each step appears benign.

Do not embed instructions in uploaded files, project text or shared content that are intended to manipulate the Service, another user's session, or our systems into acting outside their intended behaviour, exposing data, or ignoring their instructions. Do not attempt to extract system prompts or internal configuration.

Do not ask the Service to produce exploit chains, credential stealers, ransomware payloads, backdoors, or code designed to evade detection or gain unauthorised access. Defensive work, secure-coding guidance and analysis of your own systems remain permitted.

Do not use the Service to plan or produce content that impersonates a real person or organisation in a deceptive way, fabricates statements or events attributed to real people, creates non-consensual intimate imagery, or misleads people about elections, public health or emergencies.

Do not use the Service to design systems that make consequential decisions about people without meaningful human review, including credit, insurance, employment, housing, education, benefits, immigration or criminal-justice decisions, and do not use it to build social-scoring systems or applications prohibited under applicable AI regulation. Do not represent Output as reviewed professional advice; the Terms of Service explain why it is not.

Do not present AI-generated Output as human-authored where doing so would deceive someone to their detriment, or where a law, employer, client or academic institution requires disclosure.

We rely on automated signals such as rate-limit and abuse detection, provider notifications, and reports from users and third parties. We do not routinely review Customer Content; we access it to investigate a specific report, a suspected serious violation, or a security incident, as described in the Privacy Policy.

For a first, non-serious violation our usual response is a warning: we explain what we found, what needs to change, and by when. Most matters end here. We may also restrict a specific feature, remove offending content, or apply tighter rate limits while the issue is resolved.

Where a violation is serious, repeated, or ongoing, we may suspend the Account or specific features. Suspension is normally preceded by notice, except where immediate action is needed to protect users, the Service or a third party. Suspension for cause does not entitle you to a refund for the suspended period.

We may terminate an Account for a material or repeated violation that is not remedied, or immediately and without prior notice for the most serious categories:

• child sexual abuse material or child endangerment;
• credible threats of violence, terrorism or material support for it;
• distribution of malware or active attacks on systems;
• large-scale fraud, phishing or identity theft; or
• conduct that exposes us or our providers to legal liability or regulatory action.

On termination, remaining Credits are forfeited and no refund is due for the terminated period. Export your work before an enforcement deadline passes where you have been given one.

We aim to match the response to the conduct, considering severity, intent, harm caused, whether it was repeated, and how you engaged with us. Honest mistakes are treated as mistakes.

Where the law requires it, or where there is a risk of serious harm, we may preserve evidence and report conduct to law enforcement or a regulator, and cooperate with lawful requests.