PromptCreek
Sign up

Acceptable Use Policy

Last updated: August 28, 2026

This policy explains what you may and may not publish or do on PromptCreek. It forms part of our Terms of Service, and it is the document we cite when we remove something. It applies to everything you submit: prompts, titles, descriptions, instructions, example outputs, reviews, images, and profile information.

A single rule underlies the rest: a prompt is an instruction to an AI system, so a prompt that exists to make an AI do something harmful is itself harmful, whatever the wording.

1. Content that is never allowed

Illegal and dangerous

  • Content that is illegal under Romanian or EU law, or that instructs others in committing a crime
  • Prompts designed to produce working malware, ransomware, exploits, credential-harvesting pages, or phishing messages
  • Instructions for building weapons, explosives, or chemical, biological, radiological, or nuclear agents, or for synthesising controlled substances
  • Content promoting terrorism, violent extremism, or organised violence, or glorifying its perpetrators
  • Content that encourages suicide, self-harm, or disordered eating, or presents them as desirable
  • Trafficking of people, weapons, drugs, endangered species, or stolen data

Sexual content and minors

  • Any sexual content involving minors is absolutely prohibited, including fictional, drawn, animated, or AI-generated depictions, and any prompt engineered to produce it. We report such content to the competent authorities and permanently ban the account. There is no appeal.
  • Sexualised depictions of real, identifiable people who have not consented, including deepfake material
  • Non-consensual intimate imagery, or prompts intended to generate it
  • Pornographic content. PromptCreek is not a platform for adult material.

Hate, harassment, and targeting people

  • Content that attacks or demeans people on the basis of race, ethnicity, national origin, religion, disability, age, sex, sexual orientation, gender identity, or another protected characteristic
  • Harassment, threats, stalking, or intimidation of any person
  • Doxxing — publishing someone's home address, phone number, ID documents, or other private details
  • Prompts built to attack, impersonate, defame, or manufacture evidence against a specific real person or business

Deception

  • Impersonating a person, company, or brand, including in your display name or profile
  • Prompts for generating fake reviews, fake credentials, forged documents, or fabricated news
  • Coordinated political manipulation, astroturfing, or vote manipulation
  • Health, legal, or financial claims presented as professional advice when they are AI-generated
  • Scams, pyramid schemes, and prompts designed to defraud

Other people's rights

  • Copying prompts, text, or images you do not have the rights to publish
  • Reproducing substantial parts of a copyrighted work inside a prompt
  • Using someone else's trademarks in a way that suggests endorsement
  • Publishing another person's personal data — names, emails, photos, client details, employee information — without a lawful basis and their knowledge

2. Protect your own information

Prompts on the Free plan are published publicly the moment they are generated, and public prompts are indexed by search engines. Before you publish, remove:

  • API keys, passwords, tokens, and connection strings
  • Client names, contract terms, unreleased plans, and other confidential business information
  • Personal data about yourself or anyone else, including internal email addresses

We may remove content containing exposed secrets or personal data without notice, to limit the harm. That is a protective step, not a penalty. If you need prompts to stay private, use a paid plan — see Terms of Service, section 4.

3. Quality and platform integrity

These are not moral rules; they keep the library usable.

  • No spam, mass-produced duplicates, or near-identical prompts published repeatedly
  • No prompts whose purpose is advertising, link-dropping, or SEO for an unrelated site
  • No gibberish, empty, or placeholder prompts
  • No reward farming — low-effort reviews, self-reviews, duplicate accounts, or content created only to trigger a community challenge
  • No rating manipulation, whether inflating your own prompts or attacking someone else's
  • Reviews should reflect genuine use. Do not trade or sell reviews.

4. Technical conduct

  • Do not attempt to access accounts, data, or systems you are not authorised to access
  • Do not probe, scan, or test the security of the platform without our written permission
  • Do not scrape or bulk-extract the library, or use automation beyond ordinary use of the site and API
  • Do not resell, sublicense, or redistribute the library as a dataset or competing product
  • Do not circumvent credit limits, paywalls, rate limits, authentication, or other technical controls
  • Do not overload the service or interfere with anyone else's use of it
  • Do not upload malware or malicious files

If you find a security vulnerability, please report it to hello@promptcreek.com and give us reasonable time to fix it before disclosing it. We will not pursue good-faith researchers who follow this.

5. A note on jailbreak prompts

Prompts that discuss, analyse, or test the limits of AI systems are allowed — that is legitimate work. What is not allowed is a prompt whose evident purpose is to make an AI system produce content banned by section 1, or to defeat safety controls in order to cause harm. We judge by the apparent purpose and likely use, not by whether a technique is described.

6. How we enforce this

We do not review everything before it goes live. We act on reports from users, on our own checks, and on notices from third parties. Depending on severity and history we may:

  • Edit or re-categorise the content
  • Remove it from curated listings, or de-index it from search engines
  • Make it private, or delete it
  • Reverse credits or rewards obtained through abuse
  • Warn, restrict, suspend, or permanently ban the account
  • Report the matter to law enforcement where the law requires it

Most first-time issues get a warning and a chance to fix them. Content covered by the sexual-content-involving-minors rule results in an immediate permanent ban.

7. Reporting and appeals

Every prompt page carries a "Report prompt" link, which opens an email to us with the prompt already identified. You can also write to hello@promptcreek.com directly about anything on the platform. Say what is wrong — spam, inappropriate content, harassment, copyright, or anything else — and a person will review it.

If we act on your content or your account, we will tell you the main reason by email. If you think we were wrong, reply within 30 days and a person will review the decision. If we made a mistake, we will put it back. The process is described in Terms of Service, section 9.

8. Changes

We update this policy as new problems appear. Material changes are announced the same way as changes to the Terms. Filing reports in bad faith, repeatedly, is itself a breach of this policy.