Content Moderation Policy

This Content Moderation Policy describes how Wikily reviews, approves, monitors, and acts on user-generated content (“UGC”) across our gaming wiki platforms (ARK: Survival Ascended, Once Human, Palworld, Hytale and other games). It is published for the benefit of our users, our advertising and demand partners, and any other party with a legitimate interest in how content is moderated on Wikily. It supplements our Community Guidelines, Terms of Service, and Acceptable Use Policy.

This Policy was last updated on Monday 27th of April, 2026, and is the current and valid version.

1. SCOPE

   Wikily is primarily an editorial platform. The majority of content (game databases, maps, calculators, item and creature data) is produced and curated by Wikily editors. UGC surfaces are limited to:

   • Comments on building templates and other community content;
   • User-uploaded building templates and blueprints;
   • The community Discord server directory;
   • Map markers and map contributions;
   • Trading listings and trader reviews;
   • Community guides and wiki edits;
   • User profile fields (display name, bio, social links).

   This Policy applies to all of the above.

1. PROHIBITED CONTENT

   The categories of content prohibited on Wikily are set out in our Community Guidelines. In summary, we do not permit hate speech, harassment, threats, sexual content involving minors, adult sexual content, graphic violence, self-harm content, illegal activity, scams and fraud, spam, intellectual property infringement, malicious files, or unauthorized sharing of personal information.

1. MODERATION MODEL

   Wikily operates a layered moderation model combining:

   • Pre-publication automated screening on every UGC submission;
   • Human pre-moderation for higher-risk content categories;
   • Post-publication monitoring and user reporting for content published after automated screening;
   • Audit logging of every moderation decision; and
   • Account-level enforcement for users who repeatedly or seriously violate our policies.

1. AUTOMATED SCREENING
   • Every UGC submission - comments, template titles and descriptions, Discord server names and descriptions, map contributions, trading listings, wiki edits, guides, and user profile fields - is screened in real time at the point of submission.
   • We use OpenAI’s Moderation API (model omni-moderation-latest) to classify both text and images submitted to the platform. The API returns category-level scores for hate, harassment, threats, self-harm, sexual content, sexual content involving minors, violence, graphic violence, and illicit content.
   • We apply OpenAI’s default thresholds for most categories and a stricter threshold than the OpenAI default for sexual content. Submissions that exceed our thresholds are rejected and never become publicly visible.
   • In addition to AI screening, every submission passes through input validation (length limits, format checks), URL allow-listing, basic spam controls (rate limits, duplicate detection, no-link rules in comments), and Cloudflare Turnstile (CAPTCHA) bot protection on form submissions.

1. HUMAN PRE-MODERATION
   • The following content categories require approval by a Wikily moderator before becoming publicly visible: Discord server listings, map contributions, community guides, and wiki edits. Submissions enter a pending state and are not indexed or shown publicly until a moderator approves them.
   • Moderators review pending submissions through an internal review dashboard. Each decision (approve, reject) is recorded together with the moderator’s identity, a reason, the previous and new status of the content, and a timestamp.
   • Service level: pre-moderation queues are reviewed and cleared within 24 hours of submission under normal operating conditions.

1. POST-PUBLICATION MONITORING AND USER REPORTING
   • Comments, building templates, and trading listings are published immediately after passing automated screening. They remain subject to monitoring and may be removed at any time.
   • Every UGC item exposes a Report action. Reports use a typed taxonomy, including inappropriate content, spam, scam, misleading content, copyright, stolen content, dead link, and other.
   • Reports are delivered in real time to the Wikily moderation team via webhook, including the reported content, its author, the reporter, the report type, and any description provided.
   • Service levels:
     • Reports are triaged within 24 hours of submission;
     • Content found to violate our policies is removed within 48 hours;
     • Severe categories (illegal content, threats, content involving minors) are actioned immediately on detection, ahead of the standard SLAs.

1. CONTENT AUDITS
   • Real-time: automated screening runs on every submission as it is created.
   • Daily: human moderators review and clear all pre-moderation queues.
   • Rolling: user reports are triaged on a continuous basis as they arrive.
   • Periodic: Wikily conducts periodic sample audits of published UGC surfaces to verify the effectiveness of automated screening and to identify any content that should be removed.

1. ENFORCEMENT

   Where content or an account is found to violate our policies, Wikily may:

   • Remove or hide the violating content;
   • Issue a warning to the responsible account;
   • Restrict the account’s ability to post new content;
   • Temporarily suspend or permanently terminate the account;
   • Where appropriate or required by law, refer the matter to law enforcement or other competent authorities, including disclosure of account information.

   Cases involving suspected child sexual abuse material are escalated immediately and reported to the relevant authorities (including NCMEC where applicable).

1. AUDIT TRAIL
   • Every moderation decision (pre-mod approve or reject, post-mod removal, account action) is recorded in dedicated audit log tables, including the moderator’s identity, the action taken, the previous and new status of the content, and a reason.
   • Audit logs are retained for as long as the corresponding content or account exists, and for a reasonable period thereafter, in accordance with applicable law.

1. APPEALS

   Users may appeal a moderation decision by contacting [email protected] within 30 days of the action, identifying the affected content or account and explaining the basis for the appeal. Appeals are reviewed by a moderator who was not involved in the original decision where reasonably practicable, and a written response is provided.

1. DEMAND PARTNER ASSURANCES
   • This Policy is published as a public statement of Wikily’s moderation practices and may be relied upon by advertisers and demand partners assessing the brand-safety profile of the platform.
   • Additional internal documentation, including procedures for the moderation dashboard, role permissions, the audit log schema, and aggregate moderation metrics, is available to demand partners on request, subject to a customary non-disclosure agreement.

1. CONTACT

   Questions about this Policy, requests for additional information from advertising or demand partners, copyright notices, and law enforcement requests can be sent to [email protected].

1. CHANGES TO THIS POLICY

   We may update this Policy from time to time to reflect changes to our services, moderation tooling, or applicable law. The current version is always available on this page, with the “last updated” date shown above.