How Comment Reply Assistant handles your YouTube data, and the terms for using it. Written in plain language wherever the subject allows it.
Comment Reply Assistant ("the app," "we," "our") connects to a YouTube channel with the channel owner's permission and writes and posts replies to viewer comments, using an AI model to match the tone and language of each comment. It was built for the AnimiScience YouTube channel and, as it develops, may be offered to other YouTube creators who choose to connect their own channel.
This policy explains what information the app accesses through Google's APIs and elsewhere, and what happens to it. If you are a creator connecting your own channel, "you" and "your channel" below refer to you and the YouTube channel you connect.
When you connect your YouTube channel, Google asks you to approve specific permissions ("scopes"). We only request what the app needs to do its job:
| What | Why we need it |
|---|---|
| Your channel ID and channel name | To show you're connected, and to tell your own replies apart from viewers' replies in a comment thread |
| Comments on your videos (text, author name, timestamps) and any existing replies in the same thread | So the AI can read what a viewer said — and what's already been said in that thread — before writing a reply |
| A Google OAuth access token and refresh token | So the app can stay connected to your channel and post on your behalf without asking you to sign in every time |
| A log of replies the app has posted (original comment, the reply text, and when it was posted or edited) | So you can review, correct, or delete any reply after the fact |
We do not access your Google password, payment information, private messages, or anything outside the specific YouTube permissions you approve.
We use the information above only to operate the app's core, visible feature: reading comments on your videos, generating a reply with an AI model, and posting that reply to YouTube — either when you click the button yourself, or automatically on the schedule you turn on. We also use it to show you the reply log so you can edit or delete what's been posted.
We do not use your data to build advertising profiles, to train our own general-purpose AI models, or for any purpose unrelated to running the app.
In practice, that means: we only use YouTube data for the features described on this page and visible in the app itself; we don't sell it, and we don't share it with data brokers or advertisers; and any use beyond what's described here would only happen after we update this policy and, where required, ask for your renewed consent.
To write a reply, the text of the comment (and any existing replies in that thread) is sent to an AI language model — currently Anthropic's Claude, accessed through Amazon Web Services (AWS Bedrock) — solely to generate the reply text. It is not shared for any other purpose, and neither Amazon nor Anthropic is given your Google account credentials.
We do not sell, rent, or otherwise share your data with third parties for their own marketing or research purposes.
Connection tokens and the reply log are stored only for as long as your channel stays connected, on infrastructure we control (during early testing, this may be a single operator-managed computer or server; as the app grows, a hosted server with the same protections described in Section 7). Reply log entries are kept so you can review and correct past replies; you can delete individual entries yourself from the log at any time, which also removes the reply from YouTube.
You can disconnect at any time by removing the app's access from your Google Account permissions page. If you'd like your stored data deleted after disconnecting, contact us at the email below and we'll remove it.
Connections to Google and to the AI provider are made over encrypted HTTPS connections. Access tokens are never displayed in the app's interface or shared with anyone outside the systems needed to run it. As with any early-stage tool, no method of storage or transmission is 100% secure, but we take reasonable, industry-standard steps to protect your information.
This app is a tool for channel owners and is not directed at children. It processes publicly posted YouTube comments as part of replying to them, but it does not knowingly collect personal information directly from children, and it is not intended for use by anyone under the age required to hold a Google Account in their country.
If we make a material change to how the app handles your data, we'll update the effective date at the top of this page and, where required, notify connected creators directly before the change takes effect.
Questions about this policy or your data: myanimeeworldd54@gmail.com.
By connecting a YouTube channel to Comment Reply Assistant ("the app," "we," "our"), you agree to these Terms of Service. If you don't agree, please don't connect your channel.
The app reads comments on videos from a connected YouTube channel and uses an AI model to write and post replies, either when you trigger it manually or automatically on a schedule you control. Every reply is recorded in a log where you can review, edit, or delete it after it's posted.
You may only connect a YouTube channel that you own or are formally authorized to manage on behalf of its owner. You're responsible for making sure anyone else with access to your Google account doesn't use the app in a way that violates these terms.
Replies generated by this app are posted to your channel as coming from you. You remain responsible for everything posted to your channel through the app, including replies posted automatically without your prior review. The Reply Log lets you check, correct, or remove any reply after it goes live — we encourage checking in on it regularly, especially in the first weeks of using automatic checking.
Replies are written by an AI model and may occasionally be inaccurate, tonally off, or miss context a human would have caught. We do not guarantee that every reply will be factually correct, appropriate, or exactly what you would have written yourself. Do not rely on the app for replies where a mistake would carry serious consequences (for example, medical, legal, or safety claims) without reviewing them yourself first.
You agree not to use the app to:
We may suspend or end a connection that we believe, in good faith, breaches this section.
The app is currently free to use while in early testing. If we introduce paid plans in the future (for example, a trial followed by a subscription), we will describe the pricing clearly before you're charged anything, and existing users will be given notice before any change that affects them.
Because this is an early-stage tool, we don't guarantee uninterrupted availability, and features may change, be added, or be removed as it develops. We'll try to give notice of significant changes where practical.
You can stop using the app at any time by disconnecting it from your Google Account permissions page. We may also suspend or end a connection if we reasonably believe these terms have been violated, or if we discontinue the app.
The app is provided "as is" and "as available," without warranties of any kind, whether express or implied, including warranties of merchantability, fitness for a particular purpose, or non-infringement.
To the fullest extent permitted by law, we are not liable for any indirect, incidental, or consequential damages arising from your use of the app, including damages resulting from replies posted automatically to your channel. Nothing in these terms limits liability where the law does not allow it to be limited.
These terms are governed by the laws of India, without regard to conflict-of-law principles, unless local law where you live requires otherwise.
We may update these terms as the app develops. If we make a material change, we'll update the effective date above and, where required, notify connected creators before the change takes effect.
Questions about these terms: myanimeeworldd54@gmail.com.