Community Guidelines

GPTpost is a tool for publishing to other people's platforms. What you send through it lands in front of real audiences on Meta, TikTok, LinkedIn, Google, X, Pinterest, Reddit, Mastodon, and elsewhere, under our application credentials. That is why these guidelines exist and why we enforce them.

These guidelines sit alongside Acceptable Use Policy, which is the binding contractual rule set. Where the two overlap, the Acceptable Use Policy governs. This page is the plain-language version: what good use looks like, what gets you removed, and why.

1. The one rule underneath all the others

Publish as though the audience is a person, because it is.

Every other rule here follows from that. Automation is a convenience for you, not a licence to treat the people at the other end as a metric.

2. What good use of GPTpost looks like

• You publish content you have the right to publish, to accounts you or your client actually control, with the client's authority.
• You disclose commercial relationships that the law or the platform requires you to disclose: advertising, sponsorship, gifting, affiliate links.
• You keep AI-assisted content truthful, and you review it before it goes out.
• You reply to comments and messages as a human would: individually, in context, and within a reasonable time.
• You honour opt-outs immediately and permanently.
• You respect each platform's own rules, because connecting an account does not suspend them.
• You keep your team's access tidy: real accounts for real people, removed when they leave.

3. Content that is never acceptable

This list is not exhaustive, and we apply it to content published through GPTpost as well as content stored in it.

Against people

• Content that harasses, bullies, threatens, or targets an individual for pile-on abuse.
• Hate speech, or content that dehumanises or incites violence against people on the basis of race, ethnicity, national origin, religion, caste, sex, gender, gender identity, sexual orientation, disability, or serious disease.
• Content that promotes or glorifies violence, terrorism, or violent extremism.
• Content that encourages self-harm, suicide, or disordered eating.

Against children

• Any sexualised content involving a minor. There is no context in which this is tolerated. We remove it, preserve evidence, terminate the account permanently, and report it to the relevant authorities and platforms.

Illegal and dangerous

• Content that facilitates the sale of illegal drugs, weapons, endangered wildlife, stolen data, counterfeit goods, or human trafficking.
• Fraud, phishing, malware distribution, or credential harvesting.
• Content that promotes dangerous acts likely to cause serious injury.

Deception

• Impersonating a person, brand, or organisation you do not represent.
• Coordinated inauthentic behaviour: sock puppet networks, fake engagement, artificial amplification, vote manipulation.
• Fabricated media presented as real, including synthetic audio or video of a real person saying something they did not say.
• Health, financial, or civic misinformation that is likely to cause real harm, including content that suppresses or misdirects voters.

Privacy

• Publishing someone's private information without consent: home address, phone number, identity document, financial account, medical record, or private correspondence.
• Non-consensual intimate imagery.

Adult and regulated content

• Sexually explicit material, and content that breaches a platform's own adult content rules.
• Content promoting regulated goods and services, including tobacco, vaping, gambling, prescription medicines, and firearms, where the relevant platform or jurisdiction prohibits it.

4. Conduct that gets an account restricted

Content is only half of it. How you operate matters too.

• Spam. Bulk, repetitive, or near-duplicate posting designed to flood a feed or game an algorithm.
• Unsolicited outbound messaging. Sending direct messages to people who have not opted in, or outside the messaging window a platform allows. Meta's 24-hour messaging window, and equivalent rules elsewhere, apply to you when you use our connection.
• Engagement farming. Buying, selling, or exchanging follows, likes, or comments.
• Evading platform limits. Rotating accounts, proxying, or splitting activity to get around a rate limit or a restriction. A rate limit is an instruction, not an obstacle.
• Evading enforcement. Creating a new workspace to get around a suspension.
• Scraping platform data beyond what the authorised API returns, or retaining platform data past what the platform permits.
• Reselling platform data obtained through GPTpost.
• Using GPTpost to build a surveillance product, or to monitor individuals for law enforcement or intelligence purposes.
• Abusing the AI features: generating content in the prohibited categories above, attempting to extract another tenant's data through prompt injection, or using the AI to produce material that impersonates a real person.

5. Automation done responsibly

Automation is the point of the product. Doing it carelessly is what gets application credentials revoked, which harms every other customer.

Do

• Schedule content at a human cadence.
• Let approvals gate anything sensitive, and require human review of AI output before it publishes.
• Use auto-replies for acknowledgement and routing, and hand over to a person when the conversation needs one.
• Back off when a platform rate-limits you.
• Keep a person accountable for every connected account.

Do not

• Run fully autonomous public posting without a human in the loop, on any account where an error would be damaging.
• Auto-reply to every comment with the same text.
• Retry a failing publish in a tight loop.
• Present an automated reply as if it were written personally, when it was not.

6. AI-assisted content

If you use the AI features:

1. You are the author. Whatever the model wrote, you published it and you are responsible for it.
2. Check it. Models get facts wrong confidently. Review before publishing, especially anything about health, money, law, or a named person.
3. Do not use AI to impersonate a real individual, or to generate synthetic media of a real person without their clear consent.
4. Follow platform disclosure rules. Several platforms now require AI-generated or AI-altered media to be labelled. That obligation is yours.
5. Do not put other people's personal data into a prompt unless you have a lawful basis for doing so.
6. Remember the AI provider is a subprocessor of your data, listed in Subprocessors.

7. Client and agency work

If you manage accounts for clients:

• Connect an account only with the account owner's authority. Having the password is not the same as having permission.
• Tell the client which platforms are connected, what scopes are granted, and what your team can do with them.
• Disconnect promptly when an engagement ends, which deletes the tokens immediately.
• Do not move a client's data into another client's workspace.
• You are the controller of your clients' data in our contract terms. Your obligations are in Data Processing Addendum.

8. How we enforce

We act proportionately, and we prefer the smallest effective action.

We skip straight to the strongest action for child sexual abuse material, credible threats of violence, and active fraud or malware campaigns.

Platform-directed action. If a platform instructs us to remove content, disable a connection, or delete data, we comply, even if that removes content from your workspace. We have no discretion to refuse and we will tell you it happened.

9. Appeals

Enforcement decisions can be wrong. If you think yours was:

1. Email legal@oraglegpt.org within 30 days, quoting the reference in the enforcement notice.
2. Someone not involved in the original decision reviews it.
3. We aim to respond within 10 business days.
4. If we uphold the decision, we tell you which rule was breached and what, if anything, would let us restore access.

10. Reporting a problem

Include links, screenshots, timestamps, and the accounts involved. We acknowledge reports and act on them, though we cannot always tell you what action we took about someone else's account.

11. Changes

We update these guidelines as platform rules and abuse patterns change. Material changes are announced in-product and by email to workspace administrators. The date at the top always reflects the current version.

12. Related pages

GPT LLM ORAGLE Ltd. Liability Co.
30 N Gould St, Ste N
Sheridan, WY 82801
United States