Content Policy

Effective Date: April 23, 2026  |  Last Updated: April 23, 2026

1. Our Approach

The Registry is a public, community-driven catalog of reusable prompts ("atoms"), prompt compositions ("comps"), and meta-compositions ("mcomps"). We aim to support a broad range of legitimate technical, creative, educational, scientific, and commercial uses while protecting Users, the public, and the integrity of the Service from harm. We rely on a combination of automated security scanning, community reporting, and manual review to enforce this Policy.

Microbiocol AI is a private operator of an interactive computer service within the meaning of Section 230 of the Communications Decency Act, 47 U.S.C. § 230. Decisions to publish, remove, or restrict User Content are made in our sole discretion as part of operating the Service.

2. Prohibited Content

You may not publish, distribute, or facilitate the distribution of prompts that are designed, intended, or reasonably likely to:

2.1 Generate Child Sexual Abuse Material ("CSAM")

We have zero tolerance for content that produces, depicts, sexualizes, or facilitates the abuse or exploitation of minors. This includes prompts that solicit, generate, or evade safety filters around CSAM in any form. We report apparent violations to the National Center for Missing & Exploited Children ("NCMEC") as required by 18 U.S.C. § 2258A.

2.2 Produce Weapons or Mass-Casualty Instructions

Prompts that provide working instructions for the synthesis or weaponization of:

• Chemical, biological, radiological, or nuclear weapons capable of causing mass casualties;
• Improvised explosive devices designed to harm people or critical infrastructure;
• Cyberweapons designed to disable critical infrastructure (energy, water, healthcare, financial systems);
• Untraceable firearms or ghost guns in violation of the Undetectable Firearms Act, 18 U.S.C. § 922(p), or comparable laws.

2.3 Facilitate Targeted Violence or Terrorism

Prompts that promote, glorify, plan, or facilitate acts of terrorism, targeted violence, hate crimes, or incitement to violence against any person, group, or location, including content prepared for or supporting designated foreign terrorist organizations under 8 U.S.C. § 1189 or comparable lists.

2.4 Conduct Unauthorized Access or Cybercrime

Prompts designed to: gain unauthorized access to computer systems in violation of the Computer Fraud and Abuse Act, 18 U.S.C. § 1030; develop or distribute malware, ransomware, spyware, keyloggers, or wiper tools; conduct credential-stuffing, account takeover, or large-scale phishing campaigns; or evade widely deployed authentication, payment, or anti-fraud systems for unlawful purposes.

We support legitimate security research. Prompts intended for educational use, defensive analysis, capture-the-flag competitions, authorized penetration testing, or coordinated vulnerability disclosure are permitted provided they: clearly disclose their defensive or educational purpose; do not include functional zero-day exploits against systems the User does not have authorization to test; and do not target specific identified victims.

2.5 Engage in Fraud or Deception

Prompts designed to: commit financial fraud, identity theft, or money laundering; impersonate a real person, company, or government agency for the purpose of deception; generate fake reviews, fake news, or coordinated inauthentic behavior; manipulate elections or interfere with democratic processes; or generate deceptive deepfakes or synthetic media of identifiable individuals without their consent.

2.6 Produce Non-Consensual Intimate Imagery or Sexual Content of Real People

Prompts designed to generate, depict, or distribute sexual content of real, identifiable individuals without their explicit, verifiable consent ("non-consensual intimate imagery" or "NCII"), including AI-generated sexual deepfakes. This is prohibited regardless of whether the depicted person is famous, in compliance with applicable state and federal laws (including 15 U.S.C. § 6851 and analogous state statutes).

2.7 Harass, Threaten, or Stalk Individuals

Prompts designed to harass, threaten, intimidate, dox, stalk, or facilitate violence against any specific individual or small group, or to compile databases of personal information for such purposes (including in violation of 18 U.S.C. § 2261A).

2.8 Promote Self-Harm

Prompts that encourage, promote, or provide instructions for suicide, self-harm, or eating disorders. Educational, harm-reduction, clinical, and crisis-resource content is permitted when accompanied by appropriate context and safety information.

2.9 Spread Hateful Content

Prompts whose primary purpose is to dehumanize, demean, or incite hatred or discrimination against individuals or groups based on protected characteristics including race, ethnicity, national origin, religion, caste, sexual orientation, gender, gender identity, disability, or serious disease.

2.10 Infringe Intellectual Property or Privacy Rights

Prompts that:

• Reproduce or systematically extract substantial portions of copyrighted works without authorization;
• Circumvent technological protection measures in violation of 17 U.S.C. § 1201 (DMCA anti-circumvention);
• Disclose trade secrets in violation of the Defend Trade Secrets Act, 18 U.S.C. §§ 1836 et seq.;
• Disclose personal data in violation of GDPR, CCPA/CPRA, HIPAA (for protected health information), GLBA (for financial information), or other applicable privacy laws;
• Bear or imitate trademarks belonging to others in a manner likely to cause confusion or constitute false endorsement.

2.11 Spam, Quality Abuse, and Squatting

Prompts that:

• Are duplicate or near-duplicate uploads designed to manipulate search rankings;
• Exist solely to promote, advertise, or link to unrelated commercial offerings;
• Reserve namespace under reserved or trademarked codes (such as microprompt, admin, core, or third-party trademarks) without legitimate use;
• Contain hidden, encoded, or steganographic payloads intended to alter behavior in ways inconsistent with the prompt's stated purpose;
• Use prompt-injection techniques to subvert downstream model safety controls in a manner not disclosed in the prompt's title or content.

2.12 Violate Sanctions or Export Controls

Content that violates the U.S. Export Administration Regulations (15 C.F.R. Parts 730–774), the International Traffic in Arms Regulations (22 C.F.R. Parts 120–130), sanctions administered by OFAC, or comparable regimes in other jurisdictions.

3. Restricted Content

Some content is restricted but not categorically prohibited. We may permit such content only when it is clearly labeled, contains appropriate context, and is not designed for harm:

• Adult content involving consenting fictional characters — must be tagged appropriately and may not be discoverable via category browsing.
• Defensive security tooling — see Section 2.4 above for the conditions under which security research content is permitted.
• Medical, legal, or financial information — must include disclaimers that the content is not professional advice and that Users should consult licensed professionals.
• Prompts that intentionally jailbreak safety filters — permitted only when the purpose is documented research disclosure and the prompt itself does not produce content otherwise prohibited by this Policy.

4. Reserved Codes and Namespaces

Certain prompt codes are reserved for the official microprompt publisher to prevent confusion and impersonation. These include but are not limited to:

• Core system codes: core, system, base, init
• Security-sensitive codes: security, sec, auth, admin
• Prefixes: mp-, micro-, official-, core-, system-

Attempts to publish under reserved codes will be rejected. We may expand this list at any time and reserve the right to reclaim handles or codes that infringe trademarks or impersonate third parties.

5. Automated Security Scanning

All published prompts are subject to automated security scanning at the time of upload. The scanner detects patterns associated with prompt injection, credential exfiltration, malicious URL inclusion, and other high-risk constructs. High-risk content is rejected at the publish step and is not made publicly available. The scanner is not infallible, and acceptance of a prompt by the scanner does not imply our endorsement of its safety or accuracy.

6. Reporting Violations

If you believe content on the Registry violates this Policy, please report it to info@microbiocol.com. Include:

• The full prompt identifier (e.g. mp*author:code) or URL;
• A description of the alleged violation and the specific section of this Policy implicated;
• Any context or evidence that would help us evaluate the report; and
• Your contact information so we can follow up if needed.

For copyright-specific complaints, follow our DMCA notice procedure in Section 9 of the Terms of Service. For privacy-related complaints (such as exposure of personal information), contact info@microbiocol.com.

7. Enforcement

When we determine that content violates this Policy, we may take one or more of the following actions:

• Remove or restrict access to the offending content;
• Issue a warning to the publishing User;
• Suspend or terminate the User's account;
• Remove the User's other content if a pattern of violations is established;
• Reclaim the User's handle and/or block the associated public key from further publication;
• Refer the matter to law enforcement, NCMEC, or other appropriate authorities;
• Cooperate with valid legal process to disclose information about the User.

We will use reasonable efforts to provide notice to affected Users and an opportunity to appeal, except where doing so would compromise an investigation, endanger any person, or violate applicable law. Appeals may be sent to info@microbiocol.com.

8. Repeat Infringer Policy

Consistent with our obligations under the Digital Millennium Copyright Act, 17 U.S.C. § 512(i), we will terminate the accounts of Users who are determined to be repeat infringers of intellectual property rights. We will also terminate accounts of Users who repeatedly violate other sections of this Policy.

9. Government Requests

We respond to lawful government requests for information about Users. We require valid legal process (such as a subpoena, court order, or search warrant) appropriate to the type of information requested and the jurisdiction of the requesting authority. We will, where permitted by law, notify affected Users of legal requests for their information so they have an opportunity to challenge the request.

10. Modifications

We may revise this Policy from time to time. Revisions take effect upon posting. Continued use of the Service after a revision constitutes acceptance of the revised Policy.

11. Contact