Terms of Service

Last updated: June 21, 2026

By using thyfwxit.com, the Nexus AI Terminal at thyfwxit.com/nexus, or any related subdomain, you agree to these terms. The site is operated solo by Xavier Scott (THYFWX). It is a personal hobby project, not a commercial service.

What you can do

• Visit the site, play the games, chat with the AI, generate SFW images, set a leaderboard handle.
• Use the public read-only content (changelog, leaderboards, etc.).
• Sign in with Google to unlock memory, image generation, leaderboard submissions, and the Unfiltered chat mode.
• Donate via Buy Me a Coffee if you want to support hosting and AI costs.

What you can't do

• No illegal content. Don't use Nexus to plan crimes, harass real people, generate CSAM, or attack systems.
• No copyrighted, trademarked, or real-person content. Don't prompt the image generator or the AI to reproduce copyrighted characters, brand or company logos, trademarked designs, the work of a specific living artist, or the likeness of real, identifiable people. You alone are responsible for making sure your prompts and their output do not infringe anyone's copyright, trademark, or publicity rights. Outputs are AI-generated and synthetic.
• No critical-pattern abuse. Repeated prompts about weapons of mass destruction, kidnapping, body disposal, school violence, or CSAM trigger automatic 30-minute lockouts and operator alerts.
• No bots / scraping. Automated access not pre-approved by the operator is not permitted.
• No account sharing or impersonation.
• No chosen handles containing slurs, harassment of others, or impersonation. Handles are filtered server-side; offending submissions are rejected.
• No reverse-engineering for malicious purposes. Looking under the hood for fun is fine; weaponizing it is not.
• No prompt injection or jailbreak attempts. The AI treats your messages and the chat history as untrusted input. It won't follow instructions hidden in your text, reveal its own setup, or drop its safety rules no matter how you word it. Trying just wastes your own time.

The Unfiltered chat mode

Unfiltered is a candid chat mode that hedges less on mature topics. It still refuses critical-content prompts (illegal acts, CSAM, weapons synthesis, and the like). The "unfiltered" part is tone, not safety. By using it, you confirm you are 18 or older, and it requires a Google sign-in plus a one-time age confirmation.

Lockouts and bans

• Hostile/explicit prompts in non-Unfiltered modes incur strikes. Three strikes triggers an escalating lockout (30s → 30 min). Strikes auto-decay after 30 minutes of clean behavior.
• Critical content (violence toward people, CSAM, weapon synthesis, body disposal, kidnapping, hate speech) triggers an immediate 30-minute lockout across all modes plus an operator alert.
• Slurs trigger an immediate 30-minute lockout with no warning ladder.
• Repeat offenders may be permanently banned by account, IP address, and device fingerprint.
• Lockouts and bans are server-enforced; clearing browser storage, switching accounts, switching to guest mode, or changing networks does not bypass them.
• Rate limiting: the chat API is capped per minute, and guests also have a daily message limit. Guests get the tightest cap, signed-in Google accounts get more headroom, and bursts of rapid or scripted requests get a short cooldown, so one script can't burn the free AI quota everyone shares.
• Leaderboard integrity: scores are validated on the server. Impossible or out of range scores are rejected, submissions are capped per IP each day, and scripting fake scores gets your IP blocked from the leaderboard.
• Device fingerprinting: This site generates a non-identifying device fingerprint (a short hash derived from your screen resolution, GPU, timezone, and browser rendering characteristics) to enforce bans and prevent circumvention. This fingerprint cannot identify you personally. It identifies your device. It is stored only if you are banned and is used solely for ban enforcement.

AI output disclaimer: read this

AI responses can be wrong, biased, outdated, incomplete, or entirely fabricated. The AI does not know you, your situation, or your context. It generates text based on patterns, not understanding.

• Do NOT enter passwords, Social Security numbers, credit card numbers, medical records, or any sensitive personal information into the chat. Your prompts are sent to third-party AI providers (Groq, Google Gemini) to generate responses. Once sent, I have no control over how those providers process or store your input.
• Do NOT rely on Nexus for medical, legal, financial, or safety-critical decisions. The AI is not a doctor, lawyer, accountant, or safety authority. It's a language model that predicts the next word.
• Do NOT treat AI-generated images as real photographs. All generated images are synthetic. Using them to impersonate, defame, or deceive is your liability, not mine.
• AI output may contain factual errors, hallucinated references, outdated information, or biased perspectives. Always verify important claims independently.
• The operator (Xavier Scott) accepts zero liability for any actions taken, decisions made, or consequences suffered based on AI-generated text or images from Nexus.
• By using the AI features, you acknowledge that you are interacting with a machine learning model, not a human, and that you assume full responsibility for how you use its output.

Third-party AI providers

When you send a message in any chat mode, your prompt is forwarded to one or more of the following AI providers to generate a response:

• Groq: primary text generation (LLaMA models)
• Google Gemini: fallback text generation
• Replicate: paid image generation (Flux model)
• Pollinations: free image generation fallback

Each provider has their own privacy policy and data handling practices. I do not control what they do with your input after it leaves my server. This is why you should never enter sensitive personal data into the chat.

Quotas

• Image generation requires Google sign-in. 15/day for signed-in users, 150/day for premium supporters, unlimited for the operator. Guests cannot generate images.
• Quotas reset at midnight UTC.
• Quotas are subject to change without notice based on AI-provider costs.

No warranty

The site is provided "as-is" with no warranty of availability, fitness for purpose, or accuracy. The operator may take the site down for maintenance, cost reasons, or any reason at all without notice.

Limitation of liability

To the fullest extent allowed by law, the operator's (Xavier Scott's) total liability to you for any claim arising out of or relating to the site, the AI, or any text or image output, under any legal theory, will not exceed one hundred US dollars ($100) or the total amount you paid the operator in the twelve months before the claim, whichever is greater. The operator is not liable for any indirect, incidental, consequential, special, exemplary, or punitive damages, or for any lost profits, lost data, or loss of goodwill, even if told such damages were possible. Some places do not allow some of these limits, so parts of this section may not apply to you.

Termination

The operator reserves the right to terminate any account or block any IP at any time for any reason. Reasons may include but are not limited to: violations of these terms, abuse of the AI, attempts to circumvent lockouts, automated/scripted access, or operator discretion.

Intellectual property

• The Nexus AI Terminal codebase is © Xavier Scott. Open-source components are credited in the source.
• Your AI conversations and AI-generated images: you are responsible for their use. The site retains no ownership over user-prompted output.
• Don't copy the site's design and pretend you built it. If you want to fork it, ask.

Your eligibility and promises

• Age. This service is not directed to children under 13, and anyone under 13 may not use it or submit information through it. By using the site you confirm you are at least 13, and at least 18 to use the Unfiltered mode.
• Your promises. By using the AI you represent that you have the rights needed to submit your prompts, and that your prompts and their output will not infringe anyone's copyright, trademark, or publicity rights, or break any law or these terms.
• Third-party model rules. Image generation runs on outside models (Replicate's Flux and Pollinations). By generating an image you also agree to those providers' terms, which forbid, among other things, non-consensual or intimate imagery, deceptive deepfakes of real people, and use for surveillance, biometric identification, or military purposes. Generated images are for your own personal, non-commercial use.

Copyright and DMCA

To report content you believe infringes your copyright, email xavier@thyfwxit.com with your contact details, the work concerned, the address of the infringing material, a statement made in good faith that the use is unauthorized, and a statement under penalty of perjury that your notice is accurate. The operator removes infringing material promptly and bans repeat infringers. The AI creates images from your prompts using outside models, so you are responsible for making sure your prompts and their output do not infringe anyone's rights. For complaints that are not about copyright, for example that an AI-generated image resembles a real person, or content you believe is defamatory, email the same address and the operator will review and remove or address it promptly.

Indemnification

You agree to indemnify and hold the operator (Xavier Scott) harmless from any claim, damage, loss, or legal cost that arises from your use of the site, your violation of these terms, or any content you generate or submit, including images and chat you create through the AI. You alone are responsible for how you use the AI and any output you create, share, or rely on.

Governing law and disputes

These terms are governed by the laws of the State of Texas, USA, without regard to its conflict of law rules. Any dispute that is not resolved informally will be brought only in the state or federal courts located in Texas, and you consent to the jurisdiction of those courts. You agree to raise any dispute with the operator in good faith first, to resolve disputes individually, and to give up any right to join a class action, to the fullest extent the law allows.

Acceptance and assumption of risk

You use the site and its AI features at your own risk. By using the site you confirm you have read and accept these terms, including the AI output disclaimer, the clause limiting the operator's liability, and, for the Unfiltered mode, that you are 18 or older. If you do not accept these terms, do not use the site.

Changes to these terms

If these terms change, the new version is posted here and announced in the changelog. Continued use after a change means you accept the new version.

Contact

Xavier Scott (THYFWX): xavier@thyfwxit.com