Acceptable Use Policy

This policy is part of the Terms of Service. It applies to everything you put into Prompt Share and everything you do with it — including through the API. Breaking it can cost you your content, your store, or your account.

• Anything unlawful where you or your readers are.
• Material that infringes someone else’s copyright, trademark or other rights.
• Other people’s personal data, confidential information, or credentials of any kind.
• Sexual content involving minors, or content that sexualises them in any way. This is reported, not just removed.
• Content designed to harass, defame, threaten, or incite violence or hatred against people or groups.
• Prompts whose evident purpose is to produce malware, phishing pages, fraud, or instructions for serious physical harm.
• Deliberately deceptive content — impersonating a person or organisation, or passing someone else’s work off as yours.
• Spam: bulk near-duplicate packs, keyword-stuffed listings, or packs whose purpose is to advertise something else.

Public titles, summaries and descriptions are read by AI crawlers and agents. Do not craft them to manipulate those systems — instruction injection, hidden text, or anything intended to make an agent act against its operator. Public text is length-capped, stripped of control and bidirectional-override characters, and rendered as inert plain text, but attempting this is a breach regardless of whether it works.

• Probe, scan or test the security of the service without permission, beyond the vulnerability reporting below.
• Attempt to access another workspace’s data, or another account.
• Circumvent rate limits, quotas or the account gate on prompt bodies — including by creating multiple accounts to do so.
• Inflate view or copy counts, whether by scripts, multiple accounts, or coordinated activity. Ranking signals are deduplicated, capped and audited.
• Overload the service, or use it in a way that degrades it for others.
• Resell or sublicense access to the service.

Automation is welcome. Scraping is not. The difference is which door you use.

Allowed:

• The public JSON API at /api/public/v1, within its published rate limits.
• An API key on your own account, for your own automation.
• Search and AI crawlers indexing the public catalogue. They receive exactly what an anonymous human receives — catalogue metadata, never prompt bodies.
• /llms.txt and /llms-full.txt, which exist precisely so machines do not have to scrape HTML.

Not allowed:

• Scraping the HTML rather than using the documented API.
• Bulk-harvesting prompt bodies across many packs, whatever the method.
• Ignoring robots.txt or the published rate limits.
• Using multiple accounts or addresses to exceed a quota.

Depending on severity we may warn you, remove content, unpublish a store, restrict a feature, suspend an account, or terminate it. Serious cases are reported to the relevant authorities.

Where it is reasonable we will tell you what happened and why, and you may reply. Some controls are automatic — a soft rate-limit block, for example — and clear on their own.

To report content or behaviour that breaks this policy, contact us with the URL and what is wrong with it. For copyright specifically, see reporting and DMCA.

If you find a security vulnerability, please report it privately to support@degird.com before disclosing it anywhere else. Include what you found, how to reproduce it, and what you think the impact is.

We will not pursue action against good-faith research that: stays within your own accounts and test data, does not access or modify anyone else’s data, does not degrade the service, and gives us reasonable time to fix the issue before publishing.