Provider
Leonardo Antonio Carta (Nema)Controller
c/o Emperon Studio GmbH, Sachsenring 51a, 50677 Köln, Deutschland
This notice supplements the Privacy Policy for Nema Crew for the YouTube module. It does not repeat what the policy says for all modules. That includes the runtimes (Section 6), pseudonymisation and field masks (Section 7), the framework for modules (Section 11), backups (Section 17.6) and your rights (Section 18).
1. Provider and controller
This module is provided and operated by Leonardo Antonio Carta (sole proprietor), c/o Emperon Studio GmbH, Sachsenring 51a, 50677 Cologne, Germany, support@nema.im. It runs as a separate service on our server in Germany. We are the controller. If you use the module in an organisation, we process the data on behalf of the organisation under the Data Processing Agreement.
The module uses Google's YouTube API Services (YouTube Data API and YouTube Analytics API). Your use of YouTube is governed by the YouTube Terms of Service. Google is not the provider of this module. Google processes the data in your Google account and on YouTube under its own responsibility and the Google Privacy Policy.
2. What data the module processes
Permissions the module requests from Google when connecting:
youtube.force-ssl: view your YouTube channel, videos, playlists and comments, upload videos and change their details, create and fill playlists, reply to comments;yt-analytics.readonly: read YouTube Analytics reports for your channel, without revenue figures.
The module does not delete videos, playlists or comments, does not moderate comments and does not change channel settings.
We only receive data that you allowed Google to give us access to when connecting:
- when connecting, the channel's identifier, name and handle and the permissions you granted Google;
- about the channel: identifier, name, handle, description, country, creation date, link, visibility, number of subscribers (rounded by YouTube), views and videos;
- about the channel's videos, including private and unlisted ones: identifier, title, description, tags, category, visibility, scheduled publication, upload and processing status, reason for a rejection, "made for kids" setting, duration, views, likes and number of comments;
- about the channel's playlists: identifier, title, description, visibility, number of videos and date;
- comments on the channel's videos with up to five replies per thread: identifier, text, likes, times and the person who wrote it (display name and channel identifier);
- from YouTube Analytics, aggregated figures for the channel or a video, such as views, watch time, average view duration, subscribers gained and lost, likes, comments and shares, also by day, month, video, country, traffic source or device. These reports contain no data about individual viewers;
- when uploading, the video file and the details the agent provides for it.
3. What for
Exclusively for the functions of Nema Crew that you use and for the tasks that you or members of your organisation give an agent. The agent acts on your behalf with the permissions you granted Google:
- reading: show the channel, videos, playlists and comments, search comments by text or find held comments, get the status of an upload, evaluate statistics from YouTube Analytics;
- writing: upload videos, from a public https address or from the agent's workshop; change title, description, tags, category, visibility, scheduled publication and the "made for kids" setting; create playlists and add videos; reply to comments publicly as your channel.
The agent first shows every write action as a preview and only carries it out once it is confirmed. New videos and playlists are private by default, and subscribers are not notified by default.
Triggers: If you have set an agent to one of these triggers, we query YouTube regularly and start the agent with the result:
- New comment on my videos: every 15 minutes, the newest comments and replies on the channel's videos, up to 25 per query. Comments by your own channel trigger nothing.
- Upload processed or published: every 15 minutes, the status of the 15 newest uploads. The agent starts when YouTube has finished processing a new video, a video becomes public or an upload fails.
- Daily channel statistics: once a day, the figures of the day before yesterday compared with the week before, with the five most-watched videos of the day.
- Subscriber milestone reached: every hour, the number of subscribers. The agent starts when it passes a round mark.
Without an agent with the trigger, the query does not take place. In an organisation, the organisation can block such triggers for the module.
We do not use data from Google APIs for advertising, including personalised or interest-based advertising. We do not build profiles from it for other purposes, do not sell it, do not pass it on to data brokers or information resellers and do not use it to determine creditworthiness or for lending purposes.
4. Whom the data goes to
- to the module's service on our server;
- to Google, when an agent reads or writes data. If an agent uploads a video from its workshop, the file goes directly from the workshop to Google with a one-time upload address;
- as a result or event to the language model of the runtime you chose, to the extent an agent needs the data for the task (Privacy Policy Section 6). In the Nema runtime these are the service providers we specify with each request, as our processors (Sections 6.1, 15 and 16). In the Claude Code and Codex runtimes these are Anthropic or OpenAI under your own contract (Sections 6.2 to 6.4);
- to everyone who uses YouTube, to the extent an agent publishes something on your behalf: replies to comments are public, and others see videos and playlists depending on the visibility chosen. Whether YouTube notifies subscribers or the person whose comment is answered is governed by YouTube;
- to other members and agents in the channel in which an agent uses the data;
- to Apple's or Google's push service, to the extent the data is in the first 180 characters of a message we send you as a notification (Section 12);
- to the data centre operator on whose hardware our server runs and where our backups are kept, as our processor (Sections 17.6 and 20);
- to authorities only if we are legally obliged to.
If an agent uploads a video from an internet address, the module fetches the file from the server at that address. It only sends the request for the file, no data from your account.
Beyond this, we do not pass on data from Google APIs. The module uses no further service providers.
Countries outside the EU: For users in the European Economic Area and Switzerland, Google Ireland Limited (Gordon House, Barrow Street, Dublin 4, Ireland) is responsible for the Google account; for Google Workspace accounts, your organisation's contract with Google also applies. Google also processes data in the USA; Google LLC is certified under the EU-US Data Privacy Framework (Art. 45 GDPR), supplemented by standard contractual clauses. For the service providers of the runtimes, Section 16 of the Privacy Policy applies.
5. Storage and protection
- Credentials: We store the OAuth tokens encrypted (AES-GCM) in our database; renewed tokens likewise, together with the channel's identifier, name and handle. The module receives the tokens from our server with each call and does not store them itself.
- Content: The module has no database and does not store any YouTube content permanently. It fetches it anew with each request. It only keeps in memory: the channel details for five minutes, the state of the triggers (identifiers of reported comments, status of the newest uploads, subscriber count, reported days) and running uploads from an internet address with their result. It only passes the video file through and does not write it to disk. The state of the triggers is refreshed with every query, and a finished upload is deleted after one day. Whatever the module keeps in this way, it refreshes or deletes after 30 days at the latest. A restart of the module deletes all of it.
- Events and results in Nema Crew: What we store ourselves from YouTube data, we remove after 30 days at the latest. This covers the card for an event that triggered an agent (title and link), the event itself and the results of the YouTube tools in the stored history of an agent in the Nema runtime, in the steps of a run the details of YouTube tools and the thinking steps, knowledge entries from YouTube, and the jobs and records of the triggers. In their place it then says "Content removed after 30 days (required by the service)".
- What an agent makes of it, meaning its responses, messages and memories, we store like other content (Privacy Policy Sections 4 and 5).
- Transmission: The connection to Google and between your apps and our server is encrypted with TLS.
- Pseudonymisation: In organisations that have switched on pseudonymisation, the module's results and events that trigger an agent go to the language model pseudonymised, unless the organisation has switched this off for the module. For events, it can hide or replace individual fields with field masks, such as the names and channel identifiers of commenters. Both are off by default, and pseudonymisation has limits (Privacy Policy Section 7).
6. Retention, disconnecting, revoking
- Disconnecting: If you disconnect the module in the app, we delete the credentials from our database immediately. After that, no agent retrieves data from this channel any more. If an Admin or Owner connected the channel for an organisation, only the organisation can disconnect it.
- Revoking access at Google: In your Google account's security settings at https://security.google.com/settings/security/permissions you can withdraw Nema Crew's access at any time. The tokens stored with us then become invalid. We remove them from our database when you disconnect in the app or delete your account. Disconnecting in the app does not revoke the authorisation at Google. It remains there until you revoke it yourself. We therefore recommend doing both.
- What remains: What an agent has taken over from YouTube data into its messages or memories remains stored after disconnecting until you delete it (Privacy Policy Section 17.1). Videos, playlists and replies an agent published or changed on YouTube stay there.
- Deleting your account: With your account we immediately delete your credentials and content (Privacy Policy Section 17.5). Channels connected for an organisation remain until the organisation disconnects them or is deleted.
- Backups: The data, including the encrypted tokens, remains in our backups for up to about 6 months (Privacy Policy Section 17.6).
7. Legal basis
Art. 6(1)(b) GDPR, because the module carries out the connection you set up. For third parties' data in comments, such as commenters' names, channel identifiers and texts, Art. 6(1)(f) GDPR applies, limited to the task you give the agent. In organisations we process on their behalf (Privacy Policy Section 2). Your confirmation of this notice is not consent. To end the processing, disconnect the module.
8. Requirements of Google
Limited Use: Nema Crew's use and transfer of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements.
No training of AI models: We do not use data from Google APIs to develop, improve or train generalised AI or ML models, including in anonymised or aggregated form, and do not pass it on so that others can do so. In the Nema runtime, this data is only sent for processing to the operators we specify with each request. Only operators that neither store the requests nor use them for training are permitted (Privacy Policy Section 6.1). If you use the Claude Code or Codex runtimes with your own subscription, data from Google APIs that is in the conversation context also goes to Anthropic or OpenAI. They process it on the basis of your own contract. Whether they use your requests for their models depends on that contract and your settings there (Privacy Policy Section 6.4).
YouTube API Services: This module uses the YouTube API Services. The YouTube Terms of Service and the Google Privacy Policy apply. You can revoke access at https://security.google.com/settings/security/permissions. YouTube API data that the module keeps is refreshed or deleted after 30 days at the latest. Nema Crew also removes events, tool results and knowledge entries from YouTube after 30 days at the latest (Section 5).
Documents
This notice supplements the general privacy policy for this module. In the app you confirm it before connecting.