> For the complete documentation index, see [llms.txt](https://docs.onlymonster.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.onlymonster.ai/team-and-security-management/message-guard.md). # Message Guard Message Guard is an AI-powered, multi-layered message protection system designed to help your team stay compliant with platform rules. Message Guard is available for **OnlyFans**, **Fansly**, and **Telegram** platforms. Some Message Guard settings can be configured separately for each platform, while others are shared across all platforms in your organization. See [Platform Settings](#platform-settings) for details. If you want to keep protection active for only one platform, you can configure your preferences with the **Platforms** toggles in the Message Guard tab. We recommend using maximum protection across all platforms to ensure full account safety and consistent compliance standards. **Why It Matters** Maintaining compliance is critical to protect your organization’s reputation and avoid account bans or platform warnings. Whether violations happen by mistake or on purpose, Message Guard stops prohibited text content before it’s sent. **How It Works** * The system automatically scans all outgoing messages in real time. * If a message contains a prohibited word or phrase, Message Guard: * Blocks the message before it’s sent * Logs the attempted violation for team managers to review * All attempts are tracked in detail, allowing for transparency and follow-up coaching if needed. * With [**Message Guard Automation**](/team-and-security-management/message-guard-automation.md), you can also block a team member’s access to all accounts in case of repeated violation attempts. ## Overview The Message Guard protection system uses multiple AI-powered layers to safeguard your accounts and ensure compliance. Each layer can be configured to match your organization’s specific needs and risk tolerance. * **Bypass Protection** - Detects hidden or altered versions of prohibited words. * **Context Detection** - Analyzes the full context and intent of a message to catch violations that contain no prohibited words (e.g., "Maybe we could chat somewhere else?"). See the [Context Detection](/team-and-security-management/context-detection.md) article for details. * **Red List** - Blocks words and phrases that are likely to violate [OnlyFans' Terms of Service and Community Guidelines](https://onlyfans.com/terms). This includes terms associated with illegal, harmful, or non-compliant content as inferred from the platform’s public policies. * **Yellow List (OnlyMonster Curated Words)** - A research-based list of high-risk terms identified through industry analysis and platform monitoring. * **Custom List** - Your organization’s personal list of words or phrases you want to restrict, tailored to your specific strategy. * **Whitelist** - Add exceptions to allow specific words or phrases that would otherwise be blocked. * **Additional AI Protection** - Prevents messages that redirect fans to outside platforms (e.g., social media links or payment sites), helping avoid external traffic violations. * **Message Guard Automation** - Automatically blocks a team member’s access to all accounts after repeated violations. For more details, read the [Message Guard Automation article](/team-and-security-management/message-guard.md). **Protection Status** The Message Guard shows a real-time protection status based on which layers are currently active: * **Full Protection** - All critical protection layers are enabled (recommended for maximum account safety) * **Partial Protection** - Some essential layers are disabled * **Disabled** - No Message Guard layers are currently active. {% hint style="warning" %} We strongly recommend keeping Full Protection enabled to minimize compliance risks and ensure safe team operations. {% endhint %}
## Accessing the Message Guard **Activation** 1. Locate the Message Guard **Activate Now** button and switch it **ON** to activate the Message Guard 2. Configure your desired protection layers based on your organization's needs ### Platform Settings By default, the Red, Yellow, Custom, and White lists are the same for all platforms. | Setting | Configuration | Options | | ----------------------------- | ------------- | --------------------------------------------------------------------- | | **Red List**, **Yellow List** | Per platform | Turn the list on or off, and turn individual subcategories on or off | | **Custom List** | Per platform | Turn the list on or off, and create a separate list for each platform | | **White List** | Per platform | Turn the list on or off, and create a separate list for each platform | | **Bypass Protection** | All platforms | Cannot be turned off for an individual platform | | **Context Detection** | All platforms | Cannot be turned off for an individual platform | | **Additional AI Protection** | All platforms | Cannot be turned off for an individual platform | To switch between platforms when configuring the lists, use the platform selector in the **Keyword Filters** section. ### Bypass Protection Bypass Protection is an AI-powered security layer designed to detect and block attempts to disguise prohibited words using character manipulation or formatting tricks. This layer uses machine learning to intelligently analyze and catch these obfuscation patterns in real time. {% hint style="info" %} Note: Bypass Protection operates separately from other word lists. Even if a word doesn’t appear on your filter lists, this layer may block it if an obfuscated match is detected. {% endhint %} ### Context Detection Context Detection is an AI-powered protection layer that analyzes the meaning and intent of a message instead of matching individual words. It works independently from the word lists and Bypass Protection, so a message can be blocked by any of them. All violations are logged in the Message Guard Logs with the topic identified. For more details, read the [Context Detection article](/team-and-security-management/context-detection.md). ### Prohibited Words Protection The Message Guard organizes restricted language into color-coded categories for clear visibility and flexible control. Each category includes optional subcategories. #### **Red List - Critical Words** This list includes the most dangerous terms that can trigger account penalties or bans. These are high-risk words that directly violate platform policies. Recommendation: Always keep this category enabled. #### **Yellow List - OnlyMonster Curated** A comprehensive list compiled by the OnlyMonster team based on research, platform violations, and ongoing policy changes. This list is regularly updated to reflect the latest compliance standards and spot emerging violation patterns. {% hint style="warning" %} Recommendation: Strongly recommended to keep it enabled for up-to-date protection. {% endhint %} #### **Custom List** Create your own list of restricted terms that apply to your agency’s internal policies or brand requirements. Use Cases Include: * Blocking competitor names * Restricting non-approved promo links * Enforcing tone-of-voice or brand language standards #### **Whitelist - Approved Words** The whitelist allows you to specify the words that should always be allowed, even if they appear in restricted categories. **Important Considerations** * **Layout-Sensitive:** Keyboard layout matters (e.g., English "a" ≠ German "ä"). * **Language-Sensitive:** Accents and special characters create distinct entries (e.g., "Resume" ≠ "Résumé"). * **Exact Match Required:** Words must be entered exactly as they should appear, including capitalization and spelling. **When the Bypass Protection is Enabled** * Whitelisted words must be spelled fully and correctly. * Obfuscated versions (with symbols, extra spaces, numbers, etc.) will not be recognized and may still be blocked.
#### Word Categories and Subcategories The Message Guard organizes prohibited words into detailed subcategories for precise control. **Category Management** * **Toggle All Categories** - Use the Toggle All switch to enable or disable all subcategories at once. * **Individual Category Selection** - Turn protection on or off for specific subcategories on the selected platform, depending on your workflow and risk level. * **Word List Preview** - Click any subcategory to view the full list of restricted terms inside it. This granular approach allows you to enable protection for critical categories while permitting your team to use words from less restrictive categories when appropriate. ### External Platform Link Protection This Message Guard layer is designed to protect your organization by preventing fans from being redirected to outside platforms or contact methods that bypass your monetization flow. The system can detect and block: * **Website URLs** - Prevents sharing external website links * **Email Addresses** - Blocks email contact information * **Phone Numbers** - Prevents phone number sharing * **Crypto Wallets** - Blocks cryptocurrency wallet addresses You can enable or disable each protection type individually, depending on your agency’s needs. For most teams, we recommend enabling all blocks.
## Message Guard Automation Message Guard Automation lets you automatically block team members who repeatedly trigger Message Guard violations. A blocked team member loses access to all accounts until the block is removed. Automation works on all platforms together. See the [Message Guard Automation article](/team-and-security-management/message-guard.md) for details. ## Restricted Words Logs The Message Guard maintains comprehensive logs of all blocked message attempts, providing managers with full visibility into violations and potential risks.. **Log Information** Each logged event includes: * **Restricted Word/Phrase** - The prohibited word or phrase that triggered the protection * **Message** - Displayed as truncated preview and full message on hover * **Status** - Current status of the log entry (e.g., "Detected") * **Group/Topic** - The category that flagged the word (e.g., "Direct contact", "Underage") with a red indicator icon * **User** - The team member who attempted to send the message * **Account** - The creator's account on which the violation attempt was made * **Platform** - The platform on which the violation attempt was made * **Type** - Message type identifier * **Date** - Date and time of sending the message in which the violation attempt occurred **Filtering and Search** The logs interface includes powerful filtering options to help you quickly find specific events: * **Accounts** - Filter by specific creator accounts * **User** - Filter by team member * **Date** - Filter by date range * **Group** - Filter by protection method (Word categories, Bypass protection, Websites, etc.) * **Topic** - Filter by specific topic within categories * **Type** - Filter by message type **Log Statuses** Each log entry displays one of the following statuses: * **Detected** - Word was blocked and logged * **Archived** - Log entry has been archived for record-keeping **Log Management** To avoid clutter, you can archive old or resolved logs. Archived entries remain fully accessible but are hidden from the main view, allowing you to focus on recent or critical incidents.
#### User Experience When Message Guard Triggers When a message is blocked by the Message Guard, the user receives a real-time notification that the message wasn't sent.
If the user is blocked by Message Guard Automation for repeated violations, they lose access to all accounts until the block is removed. ## Message Guard Permissions **View Restricted Words** * Can only view the Message Guard settings, contents, and logs. No editing allowed. **Edit Restricted Words** * Can manage all Message Guard settings and options. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.onlymonster.ai/team-and-security-management/message-guard.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.