Privacy Policy
This policy describes what yt-autopilot does with data from a YouTube account that has been connected to it, including data obtained through YouTube API Services.
Who this concerns. yt-autopilot is a private tool with a single operator, who is also its owner and its only user. The only YouTube accounts it can access are accounts that this operator personally owns and has personally connected. It is not offered to the public, there is no sign-up, and it collects nothing from visitors to this website.
1. Use of YouTube API Services
yt-autopilot uses YouTube API Services. By connecting a YouTube account to it, the operator is also bound by the YouTube Terms of Service, and Google’s handling of data is described in the Google Privacy Policy.
2. What data the application accesses
2.1 Data accessed with a YouTube account’s authorisation
When a channel is connected, Google issues the application an access token and a refresh token for the three permissions listed below. Through those permissions the application accesses:
| Permission | Data accessed | Purpose |
|---|---|---|
youtube.upload |
Write access to upload a video file, its title, description, tags, category, privacy setting and thumbnail. | To publish videos the application has produced to the connected channel. |
youtube.readonly |
The channel’s identifier and title; for videos the application itself uploaded, their identifier, privacy status, processing status and public view, like and comment counts. | To confirm an upload completed, to verify the video’s privacy really is what was intended, and to detect an upload pointed at the wrong account before it happens again. |
yt-analytics.readonly |
Aggregate analytics for the connected channel: views, estimated minutes watched, average view duration and percentage, impressions and click-through rate. | To measure how the application’s own videos performed, so its future choices are based on results. |
The application does not request revenue or monetisation scopes, does not request the ability to manage comments, subscribers, playlists or memberships, does not request the ability to delete or modify content it did not create, and does not request access to any Google product other than YouTube. It never accesses a YouTube account that its operator has not connected.
2.2 Publicly available data about other channels
Separately from the above, and without any user’s authorisation, the application reads publicly available YouTube data using an API key: titles, publication dates, durations and public view, like and comment counts of videos that are visible to anyone, together with public channel statistics. This is used to understand what subjects are currently drawing attention. It is treated as Non-Authorized Data and is subject to the retention limit in section 5.
2.3 Data the application creates itself
Most of what the application stores is its own work product rather than anyone’s personal data: topic proposals, scripts, generated audio and images, rendered video files, quality-check results, decision records, cost records and internal logs.
2.4 This website
These pages are static. They set no cookies, run no analytics, embed no third-party scripts or fonts, and load nothing from another domain. The hosting provider that serves them may record ordinary server request information, as any web host does.
3. How the data is used
Authorised YouTube data is used for exactly two things: to publish the operator’s own videos to the operator’s own channels, and to measure how those videos performed so the application can choose better topics and formats next time. It is not used for advertising, not used for profiling anybody, not used to build a product for anyone else, and not used to train any machine-learning model.
3.1 Limited use
The application’s use of information received from Google APIs adheres to the Google API Services User Data Policy, including its Limited Use requirements. In particular: the data is used only to provide and improve the functions described above; it is not transferred to others except as required by law; it is not used for advertising; and no human reads it other than the operator, who is the owner of the data in question.
3.2 Third parties
No authorised YouTube data leaves the operator’s machine. Access tokens, refresh tokens, the connected channel’s analytics and the statistics of its videos are never transmitted to any third party.
The application does use a third-party AI text service to write scripts and propose topics. What is sent to it is limited to the application’s own prompts plus the publicly available video titles and aggregate engagement figures described in section 2.2 — data anyone can read on YouTube without logging in. No credential, no account identifier and no authorised analytics figure is included.
The application sells nothing, shares nothing with advertisers or data brokers, and has no affiliate or tracking integrations.
4. How the data is stored and secured
- Location
- A single database file and a working directory on the operator’s own computer. There is no cloud database and no multi-tenant storage.
- Credentials
- Access tokens and refresh tokens are encrypted with AES-256-GCM before being written to disk. The encryption key is held outside the database, in the operator’s environment configuration, so a copy of the database alone does not yield a usable credential.
- Access
- The application’s own control interface is bound to the local machine only, is not published to the internet, and requires an owner token for any action that reads or changes data.
- Secrets handling
- Credentials are not written to logs, reports or error messages.
No system is immune to compromise, and this one makes no claim to be. What it does claim is the above: encryption at rest, a key kept separately, no public exposure, and no onward transmission.
5. How long the data is kept
| Data | Retention |
|---|---|
| Access and refresh tokens | Held until the channel is disconnected or the grant is revoked, then deleted. An access token is short-lived and is replaced as it expires. |
| Statistics about channels that have not authorised the application (section 2.2) | Not more than 30 days. The application refreshes such figures from the API after 25 days and deletes any that reach 30 days without a successful refresh — implementing the 30-day limit in the YouTube API Services Developer Policies. |
| Analytics and statistics for the operator’s own connected channels | Retained while the channel is connected, as the authorised owner’s record of their own performance. Deleted on request (section 6). |
| Video identifiers and publication records for videos the application uploaded | Retained as a record of what was published where, including after a channel is disconnected, so that published videos remain accounted for. |
| Rendered video files and intermediate media | Deleted on a storage-retention schedule once the upload is confirmed; temporary render files are swept automatically. |
| Internal logs, audit records and cost records | Retained for operational and accounting purposes; they contain no YouTube account credentials. |
6. Revoking access, and deletion
Access can be withdrawn at any time, and the two steps below are genuinely different. Please do both for a complete withdrawal.
6.1 Revoke the grant at Google
Go to myaccount.google.com/permissions, select yt-autopilot, and remove its access. This immediately invalidates the refresh token so the application can no longer obtain a new access token, whatever it still holds locally. This is the authoritative step, and it does not depend on the application co-operating.
6.2 Delete the stored credential and data
Disconnecting the channel inside the application deletes its stored access and refresh tokens outright and drops the channel to a state in which it cannot publish. Note honestly that disconnecting inside the application does not revoke anything at Google — it deletes this application’s copy of the credential, which is why step 6.1 exists.
To have the remaining stored data about a connected account deleted — its analytics history, performance snapshots and publication records — email pljone88@gmail.com. Because the data lives in a single local database under the operator’s control, deletion is carried out directly and does not have to propagate through any third party. Records required for tax or accounting purposes, and the identifiers of videos already published, may be retained.
7. Legal bases and rights
Where data protection law such as the GDPR or UK GDPR applies, the lawful basis for processing the connected account’s data is the consent given at Google’s authorisation screen, which can be withdrawn as described in section 6, together with the operator’s legitimate interest in operating their own channels. Rights of access, rectification, erasure, restriction, objection and portability may be exercised using the contact address below. In this application’s case the data subject and the operator are the same person.
8. Children
The application is not directed at children, is not usable by anyone other than its operator, and collects nothing from children.
9. International transfers
Authorised YouTube data is processed on a single machine in the operator’s own country and is not transferred internationally by this application. Requests made to Google’s APIs, and to the AI text service described in section 3.2, are served by those providers’ own infrastructure under their own terms.
10. Changes to this policy
If the application’s data practices change — if it requests a different permission, stores something new, or sends data somewhere it previously did not — this policy will be updated before that change takes effect, and the effective date at the top of the page will change with it.
11. Contact
The operator of yt-autopilot is also its data controller and can be reached at pljone88@gmail.com.