How we interact and contribute
Be respectful
Challenge an explanation without attacking the person who wrote it. Do not post harassment, threats, hateful content, sexual exploitation, or targeted abuse. Make room for different learning levels and communication styles.
When someone makes a mistake, explain the correction patiently. Never shame a learner for asking a basic question.
Keep contributions useful
Use clear titles, choose relevant categories, and give enough context for others to respond. Avoid spam, repeated promotions, misleading claims, and duplicate posts. Do not artificially inflate activity, votes, or reputation.
State uncertainty when you are unsure. If you discover an error in something you shared, acknowledge it and provide the correction.
Learn honestly
Follow the rules that apply to your coursework and examinations. Do not use the community to obtain unauthorized answers during an assessment, impersonate another student, or submit someone else’s work as your own.
Use AI to develop understanding. Review AI-generated explanations before sharing them, and be clear about AI assistance when it matters to the discussion or your academic rules.
Respect privacy and ownership
Do not share another person’s contact details, identity documents, private messages, account credentials, or other personal information without permission. Remove identifying details from screenshots and study documents.
Share only material you created or have a valid basis to distribute. Credit sources. Do not upload leaked examination material, unauthorized copies of paid resources, malware, or files intended to harm others.
How our filter system works
We use layered screening systems to maintain a clean educational space while safeguarding user privacy.
Language screening
Text screening checks for inappropriate language using an English profanity filter with an additional starter list of Sinhala and transliterated Sinhala terms. It also checks some common attempts to disguise flagged words.
Educational anatomy terms are not automatically treated as profanity merely because they name a body part. Context and respectful use still matter.
Image screening
Uploaded images go through format validation and automated screening for potentially explicit sexual content. The system uses a locally hosted image classifier. Supported images are JPEG, PNG, and still GIF; animated images are not supported.
This classifier is one safety measure. It cannot reliably judge every educational context or detect every kind of harmful content.
Where screening happens
The app uses browser-side checks where implemented and server-side screening for protected submissions. Text and image moderation runs through a private local service, rather than sending those moderation requests to a paid generative AI provider. AI study features have a separate processing flow.
When a safety check cannot finish
If the screening service is unavailable, the affected submission is stopped and you are asked to try again later. A temporary screening error does not mean that you have broken a guideline.
PDF limitations
PDF uploads have file-size, page-count, and structural checks. These checks are not an antivirus scan or a review of every paragraph and embedded image. A PDF being accepted does not certify its safety, accuracy, or permission to share it.
The automatic AI assistance filter
A separate filter decides whether a post or reply needs automatic AI assistance. It can skip greetings, acknowledgments, and simple resource requests. This keeps automatic responses focused on likely academic content.
That decision is separate from content moderation. A message can be permitted in the community without receiving an AI response.
If your content is blocked
Read the message shown in the app. Review your wording or choose an appropriate educational image. Do not repeatedly disguise blocked material to bypass the checks.
If you believe the result is mistaken, keep the visible error and a description of what you were trying to share for any follow-up through an available official contact channel. Do not repost harmful content publicly to demonstrate the problem.
Community safety & enforcement
⚖️ Fair & Accountable Moderation
Content or account access may be restricted to address misuse, protect the community, or respond to applicable requirements, as described in the Terms and Conditions. Automated filters do not replace your responsibility to follow these guidelines.
If you encounter harmful material, avoid amplifying it. Preserve the post link and relevant context for reporting through an available official channel, without copying private information into a public discussion.
Contribute with care
Ask clearly, explain generously, protect privacy, and leave room for correction. The most useful contribution helps someone understand a little more than they did before.