# autotube.pro documentation Source: https://autotube.pro/docs --- # Overview URL: https://autotube.pro/docs Section: Getting started > What autotube.pro does, how these docs are organized, and where credits fit in. autotube.pro turns a video title and a channel into a finished YouTube video. You write or import a script, generate a voiceover from it, build the visuals scene by scene, render the video in the cloud, then make a thumbnail and a description. Every step is editable before you move on and can be rerun on its own, and everything the run produces — script, voiceover, images, renders, thumbnail, description and tags — is saved inside one project you can reopen at any time. ## How these docs are organized - **[Getting started](/docs/quick-start)** — signing in, your first channel, your first rendered video, [using your own script or voiceover](/docs/use-your-own), and [where everything lives](/docs/navigation). - **[Creating a video](/docs/pipeline-overview)** — the pipeline steps in order: script, voiceover, visuals, render, description, and the [presets and one-shot dialog](/docs/presets) that run them for you. - **[Channels & research](/docs/channels)** — channels hold your defaults; competitors and ideation tell you what to make next. - **[Tools](/docs/thumbnails)** — [Thumbnails](/docs/thumbnails), the [Video Editor](/docs/video-editor) and the [AI Playground](/docs/ai-playground). - **[Projects](/docs/projects)** — where a video's assets live and how to get back to them. - **[Integrations](/docs/youtube)** — connecting a YouTube channel and bringing your own API keys. - **[Account](/docs/plans-and-billing)** — plans, billing, credits and account settings. - **[Help](/docs/support)** — what to do when a run fails or something is missing. ## Credits, in one paragraph Work that calls a model — research, script writing, voiceover, images, video clips, descriptions — is paid for in credits, and your balance sits in the top bar next to the coin icon. Before anything paid runs, the app shows an **Estimated cost** in credits and asks you to confirm, and you are charged for actual usage rather than the estimate. Plans that include a daily render allowance cover rendering separately from credits. See [Credits and usage](/docs/credits-and-usage) for how a run is priced and where to see what you spent. --- # Quick start URL: https://autotube.pro/docs/quick-start Section: Getting started > From signing in to a rendered video, in the order the app expects you to work. This is the shortest path to a finished video. Each step saves into the same project, so you can stop after any of them and pick it up later from the [Projects page](/app/projects). ## Make your first video 1. **Sign in.** Sign-in is Google only, at the [sign-in page](/app/login) — no password. 2. **Create a channel.** On the [Channels page](/app/channels), click **New channel** and describe the niche. A channel holds your defaults — script style, voice, visual style, render look — so the tenth video matches the first. See [Channels](/docs/channels). 3. **Write the script.** On the [Script screen](/app/script-generator), stay on **Generate with AI**, enter a **Title**, pick the **Channel**, set a **Word count**, and click **Generate script**. Already have a script, or a recorded voiceover? Switch to **Import your own**; see [Use your own script or voiceover](/docs/use-your-own). > Tip: the confirmation dialog shows an **Estimated cost** in credits before anything runs, and its overflow menu offers **Generate Outline** if you want to check the structure first. 4. **Generate the voiceover.** On the [Voiceover screen](/app/voiceover), choose the project, pick a voice under **Voice settings**, and click **Generate voiceover**. The narration becomes the clock for everything after it. See [Voiceover](/docs/voiceover). 5. **Build the visuals.** On the [Visuals screen](/app/visuals), the app splits the narration into scenes, writes an image prompt for each, and generates them. Review and regenerate anything you do not like before moving on. See [Visuals](/docs/visuals). 6. **Render.** On the [Render screen](/app/render), pick the subtitle and layout settings and start the render. It runs in the background and the finished file lands in the project. See [Render](/docs/render). 7. **Thumbnail.** Open **Thumbnail Generator** from the **Tools** menu to generate options for the project, and the **Thumbnail Designer** to lay one out by hand. See [Thumbnails](/docs/thumbnails). 8. **Description and tags.** Open the project on the [Projects page](/app/projects) and use **Generate** in the **Description & tags** section, then copy them into YouTube. See [Description & publish](/docs/description). ## Save your presets as you go Every step has a preset control at the top of its settings panel. When a run comes out the way you want it, click the save icon, name the preset, and tick **Set as default for this channel**. That preset is then pre-selected for every future video on that channel, and it is what the **One-shot video generation** dialog — the wand icon in the top bar — uses to run the whole pipeline unattended. See [Pipeline overview](/docs/pipeline-overview). Presets are per step: script, voiceover, visuals, render and description each have their own, and each can be the channel default independently. The full picture, including the one-shot dialog, is in [Presets & one-shot generation](/docs/presets). --- # Use your own script or voiceover URL: https://autotube.pro/docs/use-your-own Section: Getting started > Bring a script you already wrote, a voiceover you already recorded, or both, and run the rest of the pipeline on it. Nothing in autotube.pro requires the AI to write or narrate for you. A pasted script and an uploaded recording go through visuals, render, thumbnail and description exactly like generated ones. ## Your own script 1. Open the [Script screen](/app/script-generator) and switch to the **Import your own** tab. The **Projects** page has the same entry under **Import your own script**. 2. Enter a **Title**, pick the **Channel**, and paste the script into the text box. It is used exactly as written: no rewriting, no trimming. 3. Click **Create project**. Importing a script costs no credits. The project opens like any other, ready for [Voiceover](/docs/voiceover). ## Your own voiceover You can bring a recording at two points. - **With the script, in one go.** On the **Import your own** tab, open **Upload voiceover (optional)** below the script box, drop the audio file in, and the button becomes **Create project with voiceover**. - **Later, on an existing project.** Open the project's assets from the [Projects page](/app/projects), find the **Voiceovers** section and click **Upload**. This works whether the project's script was generated or imported. Either way the file must be MP3, M4A, WAV, AAC, OGG or FLAC. The upload area states the size and length limits. ## Timing: transcription or a subtitle file The pipeline needs word timings to cut scenes and draw subtitles, and a plain recording has none. You choose how they are made: - **Let the app transcribe.** Leave the subtitle slot empty. The upload shows an **Estimated transcription cost** in credits; transcription runs in the background after the project is saved, and the project is ready for [Visuals](/docs/visuals) when it finishes. - **Bring an .srt.** Add a subtitle file under **Optional subtitle file** and transcription is skipped and not billed. The app warns that a low-quality file, for example one without punctuation, can make scenes line up badly; if in doubt, let it transcribe. ## What stays the same An uploaded voiceover is the project's clock from then on, exactly as a generated one would be. Scenes, subtitles and the render all follow it. You can still generate an AI voiceover for the same project later and pick whichever one you want at the visuals step. Next: [3 · Visuals](/docs/visuals), or the full [Pipeline overview](/docs/pipeline-overview). --- # Home & navigation URL: https://autotube.pro/docs/navigation Section: Getting started > What each group in the top bar does, and how to get back to a project and its assets. The top bar is the whole app. It holds the four pipeline steps, your channels and research screens, the standalone tools, and your account. ## The top bar - **Create** — the pipeline, in order: **Script**, **Voiceover**, **Visuals** and **Render**. Clicking **Create** itself opens the [script step](/app/script-generator). - **Projects** — every video you have started, at the [Projects page](/app/projects). See [Projects](/docs/projects). - **Channels** — **Channels** at the [Channels page](/app/channels) and **Competitors** at the [Competitors page](/app/competitors). See [Channels](/docs/channels). - **Ideation** — the [Ideation page](/app/ideation), title ideas built from your channel and its competitors. - **Tools** — **Thumbnail Generator** and **Thumbnail Designer** (see [Thumbnails](/docs/thumbnails)), plus **AI Playground** and **Resources**. - The wand icon opens **One-shot video generation**, which runs every step you tick from a single dialog. See [Presets & one-shot generation](/docs/presets). On a narrow screen the same entries sit behind the menu icon, with the pipeline steps and the thumbnail screens grouped under a **Tools** heading. ## Credits, account and help Your credit balance shows next to the coin icon; clicking it opens the credits and purchase dialog. The avatar menu holds **Swipefile**, **Account & Settings**, **Buy Credits**, [**Plans & Billing**](/app/billing), **Support** and **Sign Out**. Two icons sit beside it: Discord, and a question mark that opens the in-app help with a **Quick Start** tab and a **Video Tutorials** tab. ## Getting back to a project A project is one video's workspace, and everything a run produces lands in it. Open the [Projects page](/app/projects), search or filter by channel, then click the project. - The project row opens an assets panel holding the **Script**, the renders, the thumbnail propositions, the generated media — images and voiceovers — and **Description & tags**. - Clicking through to the project itself opens its script editor. The header carries the next steps — **Voiceover**, **Visuals**, **Render** and **Thumbnail** — and each one opens that step with this project already selected. Projects whose jobs are still running appear in their own section above the list, so you do not have to sit on a step page and wait. --- # Pipeline overview URL: https://autotube.pro/docs/pipeline-overview Section: Creating a video > The five steps of a video, what each one produces, and how to redo one without redoing the rest. A video is made in steps, each one feeding the next, all of them saved in the same project. You can walk the steps by hand from the **Create** menu, or tick them all in the **One-shot video generation** dialog behind the wand icon and let the run finish on its own. ## The steps in order 1. **[Script](/docs/script)** — a title, a channel and a word target become a researched, sectioned script. Produces the script text, and the project itself. 2. **[Voiceover](/docs/voiceover)** — the script is narrated. Produces the audio, plus subtitles and word timings measured against that audio. 3. **[Visuals](/docs/visuals)** — the narration is split into scenes, each scene gets a prompt, and each prompt becomes an image, a clip or a stock shot. Produces the scene list and its media. 4. **[Render](/docs/render)** — audio, scenes and subtitle styling are combined. Produces the finished video file. 5. **[Description & publish](/docs/description)** — the script becomes a YouTube description and tags. Produces the text you publish with. ## What carries forward Each step consumes the step before it, which is why the order is not optional: - The **voiceover** reads the script, so editing the script after narrating means narrating again. - The **word timings** from the voiceover decide where scenes start and end and where subtitles fall. Nothing is aligned by hand. - The **render** needs both the audio and the scenes; it applies subtitles, motion and transitions from your render settings rather than from a timeline you edit. [Thumbnails](/docs/thumbnails) are the exception: a thumbnail needs neither the voiceover nor the render, so you can make one as soon as the project exists. ## Presets, one per step Script, voiceover, visuals, render and description each have their own preset, saved from the preset control at the top of that step's settings. A preset can be set as the default for a channel, in which case it is pre-selected on that step and used by the one-shot run. Configure a channel once and later videos need only a title. Step by step: [Presets & one-shot generation](/docs/presets). ## Redoing one step Nothing is locked. Open the project from the [Projects page](/app/projects) and use the **Voiceover**, **Visuals** or **Render** buttons in its header to re-enter that step with the project already selected; the new result is added to the project alongside the old one. The one-shot dialog works the same way from the other direction: for any step you can select an existing script, voiceover or set of visuals instead of a preset, and only the remaining steps are run. Start at [Script](/docs/script). --- # Presets & one-shot generation URL: https://autotube.pro/docs/presets Section: Creating a video > Save each step's settings as a preset, make it the channel default, and run the whole pipeline from one dialog. A preset is one step's settings, saved under a name. Set one preset per step as a channel's default and every new video on that channel starts from them, which is what lets the **One-shot video generation** dialog run a whole video unattended. ## Save a preset Each step's settings panel — [Script](/docs/script), [Voiceover](/docs/voiceover), [Visuals](/docs/visuals), [Render](/docs/render) and [Description](/docs/description) — has a preset control at the top: a dropdown of your saved presets and a save icon. 1. Configure the step the way you want it and, ideally, run it once to check the result. 2. Click the save icon (**Save current settings as new preset**). Name it in the **Save Preset** dialog. 3. Tick **Set as default for this channel** if this should be the channel's default. If the channel already has one, the label says **(replaces existing default)**. Selecting a preset from the dropdown loads its settings into the form. With a preset selected you can also update it in place with the current settings, rename it, delete it, or open its details. The channel's default is marked **Default for this channel** in the list. Each step caps how many presets you can keep; the save icon tells you when you have hit it. ## Set the channel defaults Defaults can also be assigned in one place. Open a channel from the [Channels page](/app/channels), click **Settings**, and open the **Default Presets** tab. It has one dropdown per step — **Script**, **Voiceover**, **Visuals**, **Render** and **Video Description** — each listing your saved presets or **None**. Click **Save default presets**. See [Channels](/docs/channels). With defaults set, opening any step with that channel selected pre-selects its default preset. You can still pick another preset or change the settings for one video without touching the default. ## One-shot video generation The wand icon in the top bar opens **One-shot video generation**, which creates a project and runs the steps you switch on, in order, as one background job. - **Title**, **Channel** and a **Word count** (for example `15k` or `12k-18k`) are the inputs. **+ Advanced script options** adds requirements or context for the script. - Choosing a channel fills each step's preset selector with that channel's default. Steps without a usable preset stay off and say why, for example **Create a video description preset first**. - **Script** always runs. **Voiceover**, **Visuals**, **Render** and **Description** are switches; a step can also reuse something the project already has instead of a preset, such as an existing voiceover or render, when you pick an existing project. - The dialog shows the estimated credit cost before you start. Progress then appears in the dialog and on the project, and each finished step lands in the project like a manual run. The MCP tools' full video creation uses the same channel defaults, so all five must be set before an assistant can run a video for you. See [MCP server & AI skills](/docs/mcp-and-skills). ## A good order to do this in Make your first video by hand, saving a preset at each step you are happy with and ticking the default box as you go, as the [Quick start](/docs/quick-start) suggests. From the second video on, the wand does the work. --- # 1 · Script URL: https://autotube.pro/docs/script Section: Creating a video > Generate a researched long-form script from a title, or import one you already have. The script is the source of everything after it: the voiceover reads it, the scenes are cut from it, the description is written from it. The [Script screen](/app/script-generator) has two tabs — **Generate with AI** and **Import your own**. ## What you fill in - **Title** — the video title. The lightbulb button finds topic ideas for the selected channel; the bookmark button lists video titles you have saved. See [Ideation](/docs/ideation). - **Word count** — a target, not a guarantee. Accepts a single number or a range, written as `15k` or `12k-18k`. Your plan sets the ceiling; see [Plans & billing](/docs/plans-and-billing). - **Channel** — which channel's style and defaults to write in. See [Channels](/docs/channels). - **Additional instructions (optional)** — extra context, constraints or requests passed to the writer. - **Quality** — **Fast** or **Good**, the two-way trade between cost and writing quality. - **Research** — tick **Research this topic before writing** to have the topic researched on the web first and written from the sources. It adds separate search and processing charges, and the sources are kept with the project. - **Style** — under the collapsible **Style** section, teach the voice either from a **Sample script** you paste or a **Style guide** you write. The guide is passed straight to the writing model. Turning on the **Advanced** switch adds model choice and outline controls: **Outline mode** (**Auto**, **Manual** or **Disabled**), the number of sections, and a second, cheaper model for the sections after the opening ones. ## Outline first Long scripts drift. Instead of writing straight through, the app plans an outline of sections with a word budget each, then writes section by section and carries a summary forward, so the ending still knows what the opening promised. Click **Generate script** and a confirmation dialog shows the target length and the **Estimated cost** in credits. Its overflow menu offers **Generate Outline (to preview the script structure first)** — take it if you want to see the plan before paying for the whole script. You then get **Edit Outline**, **Regenerate outline**, and **Generate script with this outline**, which writes exactly the outline you approved. An outline made for a title you have since changed is set aside rather than used, and the page offers to restore that title or discard the outline. ## Editing, importing and presets The finished script appears below the form, editable in place and auto-saved, with **Export**, a copy button and **Generate Voiceover** to move on. It is also now a project, so you can leave and reopen it from the [Projects page](/app/projects). **Import your own** takes a title, a channel and the pasted script, used exactly as written, and creates the project with **Create project**. You can attach a voiceover you already have at the same time under **Upload voiceover (optional)**. Step by step, with the audio formats and the transcription choice: [Use your own script or voiceover](/docs/use-your-own). When the settings are right, save them: the preset control above the form writes a **Script preset** you can set as the channel default, so the next video starts pre-filled. Next: [Voiceover](/docs/voiceover). --- # 2 · Voiceover URL: https://autotube.pro/docs/voiceover Section: Creating a video > Turn the script into narration, and with it the timing that every later step runs on. The [Voiceover screen](/app/voiceover) reads a project's script aloud. The narration is the clock for the rest of the pipeline: scene boundaries, visuals and subtitles are all timed to it, so this step comes before visuals. Already have a recording? Skip this screen and upload it instead: [Use your own script or voiceover](/docs/use-your-own). ## Pick the project The **Project** panel has a **Choose a project** dropdown, with your newest projects as chips underneath and a **Show script** button to check what will be read. If you have text that is not a project yet, **Create voiceover from text** takes a paste and makes one. ## Choose a voice **Voice settings** opens on **Standard** and can switch to **Voice Clone**. - **Standard** — pick a **Provider** (**ElevenLabs** or **Chatterbox**, which is English only), then a **Voice**. **Select voice** opens the library: search by name or description, filter by **All**, **Female** or **Male**, and play a sample from any row. ElevenLabs also exposes a **Model** choice, and a **Manual voice ID (optional override)** if you want a voice that is not in the list. - **Voice Clone** — pick a **Clone provider** and one of your clones. **Add voice clone** opens **Add Voice Clone**: give it a **Name** and upload one clear **Reference Audio** sample, within the length and file-size limits the dialog states. You can hold a small number of clones at once, rename or delete them, and test one on a short line before committing to a full run. ### Which to use For most channels we recommend a voice clone: it is far cheaper per minute than a standard voice, and it is what many large faceless channels run on. The trade-off is that cloning takes some experimenting. The model suits some voices better than others, and the sample matters as much as the voice. If a clone sounds off, try a different reference recording first, then a different voice altogether, and use the short test line before spending on a full run. Standard voices cost more but give a predictable result on the first attempt, which makes them the safer pick for a one-off video. **Speed** is a slider around a default of 1.0x. On ElevenLabs, **Advanced voice settings** adds **Stability**, **Similarity**, **Style exaggeration** and **Enable speaker boost**, with **Reset to defaults** to undo the lot. With an ElevenLabs key stored in **Account & Settings**, **Use own key** bills your ElevenLabs account directly instead of spending credits; see [Credits and usage](/docs/credits-and-usage). ## Options before you run - **Speak chapter titles** — reads later section titles aloud. Only appears when the script has more than one. - **Loudness normalization** — evens out volume across a long narration. It adds a stated percentage to the cost, and the app suggests it afterwards if a voiceover came out quiet. - **Limit to first N words** — generates only the opening words, for testing settings cheaply. - **Open visuals generation page once finished** — jumps straight to the next step. **Generate voiceover** opens a confirmation showing provider, voice, speed, word count and the **Estimated cost** in credits. A voiceover task cannot be cancelled once started, and a failed one is refunded. ## What you get Generation runs in the background — leave the page, or follow it from **View voiceover jobs**. When it lands you get **Voiceover ready** with a player, **Download Audio** and **Continue to Visuals**. The audio, its subtitles and word timings are saved to the project and kept for the number of days the card states, so download a copy if you need it long-term. Save the settings as a **Voiceover preset** and set it as the channel default to skip this configuration next time. Next: [Visuals](/docs/visuals). --- # 3 · Visuals URL: https://autotube.pro/docs/visuals Section: Creating a video > Split the narration into scenes, write a prompt for each, and fill them with images, clips or stock. The [Visuals screen](/app/visuals) turns a finished [voiceover](/docs/voiceover) into pictures. It has two views: **Generate visuals**, where you configure the run, and **Generated scenes**, where you review what it planned and generate it. ## Settings ### Source and models **Source** picks the **Project** and which **Voiceover** to use. **Models** sets the **Default image model**, with its per-image credit cost shown, and the **Prompt model** that writes the prompts. ### Style & content **Visual guidance** is a free-text field describing how images look and what they show — there is no fixed art-style menu. **Browse styles** opens the **Style library**, a dozen proven looks for faceless channels; **Use this style** replaces your guidance and image model with that style's. **Generate from images** builds a style from reference images you upload instead. ### Pacing **Simple** takes a **Number of scenes (optional)** and a **Minimum scene duration (seconds)**, and makes every scene an AI image. **Advanced** opens a timeline you cut into segments, each with its own scene count and its own mix of **AI image**, **AI video** and **Stock footage**, plus a toggle that lets the app draw custom motion graphics for standout moments. A segment set to zero scenes holds the previous visual through that stretch. ### Characters Turn on **Characters** to keep a recurring cast consistent across scenes and videos. **Generate Characters** reads the script for recurring characters; **Upload Custom Character** adds one from an image. Each has an editable appearance, a **Role in the video**, and a presence of **AI decides** or **Always**. Your plan caps the roster size and how many characters one scene can hold. ## Prompts and scenes The footer button — **Generate image prompts**, with the scene count in it — splits the voiceover into scenes and writes one prompt per scene, then moves you to **Generated scenes**. Each card shows the scene number, its timestamps, who is cast in it, the narration it covers and its **Image prompt**. Copy the prompt, edit it inline, or use the sparkle to regenerate it with a **What should change?** instruction; the scene keeps its current image until you regenerate that separately. The cast badges open **Characters in this scene**, which rewrites the prompt to match who you pick. ## Generating and moving on Generate two or three scenes first to check the look. **Generate all visuals** then shows how many images, clips and stock shots the run needs and the **Estimated charge**, and warns that generation cannot be cancelled once it starts. Individual scenes keep their own **Generate image** or **Regenerate** button with a model menu beside it, and anything that failed gets **Retry** or **Retry all failed**. **Generation status** reports progress while it works, and **Continue to render** appears when the run is done. Generated visuals are stored with the project for the number of days the page states. Save the whole configuration — look and pacing together — as a preset from **Presets** in the **Models** header, and tick **Set as default for this channel**. Next: [Render](/docs/render). --- # 4 · Render URL: https://autotube.pro/docs/render Section: Creating a video > Combine the voiceover, the scenes and your styling into one publish-ready MP4. The [Render screen](/app/render), **Render video**, turns a project into a finished file. It runs on our servers, so you can close the page while it works. ## Source The **Project** dropdown lists only projects with a finished script, a completed voiceover and completed visuals — if yours is missing, one of those is not done. Where a project has more than one set of results, pick which **Visuals** run and which **Voiceover** to use. ## Render settings There are no named templates; the look of the video comes from these settings, which you save as a preset. - **Frame rate** — 24, 25 or 30 fps. - **Resolution** — 720p or 1080p. 1080p costs more, and the control says by how much. - **Image motion** — **Off**, a slow **Zoom/pan** on stills, or **Reveal**. - **Transition** — crossfade length between scenes, and which transitions to draw from. Below them, five switchable sections: **Subtitles**, **Text overlay**, **Image overlay**, **Effect overlay** and **Music**. Each is off until you turn it on. ## Subtitles Turn on **Subtitles** to burn captions in, timed from the voiceover's word timings. If the selected voiceover has no subtitle file, the switch is disabled and says so. - **Font** — family, weight, italic and colour. - **Size** — a slider in reference pixels, with the sample strip showing **Actual size** or **Fit**. - **Styling** — **Outline** and **Shadow** thickness. - **Effect** — **None**, **Color change**, **Hormozi emphasis**, **Highlight pill**, **Pop** or **Word fade in**. Styles that highlight a word take a highlight colour; **Word fade in** adds a **Fade time**. Subtitle placement is handled by the renderer and is not a setting. ## Music and overlays **Music** plays a track from the start of the video. Use **Upload Music** or **Use Existing Music** to choose one — tracks are your own uploads, there is no built-in library — then set **Volume** and, for a track shorter than the video, **Loop music**. The overlay sections take an uploaded image, text or looping video effect laid over the whole render. ## Running it and getting the file **Render video** opens **Confirm Render**, which shows the duration, the output settings, an optional **Duration limit (seconds, optional)** for a test cut, and the **Estimated cost**. Plans with a daily render allowance show how much of it is left; 1080p counts double against it, and past the allowance a render bills credits. You can have one render in progress at a time. Progress shows on the page, but leaving is safe. When it finishes you get a preview, **Download video**, and **Render Again** if you want another pass. The MP4 also appears in **Latest renders** at the bottom of the page and in the project's **Renders** section on the [Projects page](/app/projects), with its resolution, frame rate, duration and size — kept for the number of days the page states, so download anything you want to keep. Save the settings as a render preset and set it as the channel default. Next: [Description & publish](/docs/description). --- # 5 · Description & publish URL: https://autotube.pro/docs/description Section: Creating a video > Generate the YouTube description and tags from the script, then publish the file yourself. The last step writes the text that goes under the video. Publishing itself happens on YouTube: autotube.pro does not upload for you. ## Generate the description and tags There are two ways in, and both open the same dialog: - On the [Render screen](/app/render), the **Description & Tags** button, which appears while a render is running and again beside **Download video** when it finishes. - On the [Projects page](/app/projects), the **Description & tags** section of a project's assets panel, using **Generate**. In **Generate description & tags** you can shape the output before running it: - **Description template** — the layout of the finished description. Insert placeholders with the chips above the field: `{videoTitle}`, `{generatedDescription}`, `{chapters}`, `{sources}` and `{attribution}`. `{sources}` can be limited to a number of sources; `{attribution}` becomes the credits for any stock footage in the video ("Video by Jane Doe on Pexels" with a link) and stays empty when there is none; it is not part of the default format, so add it if you want credits in your descriptions. Leave the template empty for the default format. - **Description requirements** — anything the writer should or should not do. - **Generate YouTube tags** — on by default, with its own requirements field. **Generate description & tags** shows the credit estimate on the button. The results come back as **Generated description** and **Generated tags**, both editable in place, with copy buttons. **Regenerate** tries again; **Save to project** stores them. Saved text stays editable: reopen the project's **Description & tags** section and use **Edit**. Both fields show a character count against their cap as you type. Save the template and requirements as a description preset and set it as the default for the channel, so the next video's description comes out in the same shape. See [Channels](/docs/channels). ## Publishing There is no upload or scheduling screen in autotube.pro. To publish: 1. Download the MP4 from the render page or from the project's **Renders** section. 2. Copy the description and tags from **Description & tags**. 3. Upload the file to YouTube and paste them in, with the thumbnail you made in [Thumbnails](/docs/thumbnails). Connecting a YouTube channel to autotube.pro is for research, not publishing: it brings in your channel's video performance so [Ideation](/app/ideation) and competitor analysis have something real to work with. See [YouTube](/docs/youtube). That is the full pipeline. Back to [Pipeline overview](/docs/pipeline-overview). --- # Channels URL: https://autotube.pro/docs/channels Section: Channels & research > A channel stands for one niche and holds the defaults, competitors and YouTube link every project for it inherits. A channel is a brand with its own projects, videos and settings. It holds the defaults new projects start from, the competitors you track, and the link to the YouTube channel you publish to. ## Create a channel On **Channels**, click **New channel**. Give it a **Channel Name** and a **Color** — the colour is how the channel is marked in every list, filter and pill elsewhere in the app. You can paste one **Competitor Channel** URL while creating it; more can be added later. ## Edit the defaults Open a channel and click **Settings**. The **General** tab holds the name, the colour, the **YouTube Channel ID or URL** and the **Channel brief**. The brief describes what the channel is about, who it is for and what kinds of videos fit; AI reads it when ideating topics and titles. **Generate** writes one from the channel's top videos and uses credits; you can also write it yourself. **Save settings** applies the tab. The **Default Presets** tab sets one saved preset per pipeline step: | Field | Used by | | --- | --- | | Script | [Script](/docs/script) | | Voiceover | [Voiceover](/docs/voiceover) | | Visuals | [Visuals](/docs/visuals) | | Render | [Render](/docs/render) | | Video Description | [Description & publish](/docs/description) | Each is a dropdown of your saved presets, or **None**. Click **Save default presets**. How presets are made and what the defaults unlock: [Presets & one-shot generation](/docs/presets). Full video creation driven by an AI assistant needs all five set — see [MCP server & AI skills](/docs/mcp-and-skills). ## Competitors, YouTube and members A channel's own page lists its **Competitors** with an **Add competitor** button, and — if you own the channel — a **Members** section. See [Competitors & outliers](/docs/competitors), [YouTube](/docs/youtube) and [Collaborators](/docs/collaborators). ## Archive and delete From the row menu on **Channels**, **Archive channel** hides a channel from filters and pickers without changing any data; **Unarchive channel** brings it back. **Delete** removes the channel and all its competitors permanently. Projects attached to it are disconnected but stay in your account. --- # Ideation URL: https://autotube.pro/docs/ideation Section: Channels & research > Generate ranked title ideas for a channel from the videos that already work on it and on the competitors you track. **Ideation** turns channels into titles. It analyses real YouTube videos — your own and your competitors' — and proposes titles for the channel you picked. The segmented control at the top of the page switches between **Generate ideas** and **Competitor outliers**. ## Generate ideas 1. Pick a **Channel**. Its saved competitors appear as chips under **Sources**, along with the YouTube channel it is connected to. 2. Select the sources to analyse. Paste any YouTube channel URL and click **Add channel** to add one that is not saved. Up to five sources at a time. 3. Choose the **Idea types**: **All-time popular** draws on each source's historical best performers, **Trending** on recent videos that performed well. At least one must be ticked. 4. Set **Number of ideas** — 5, 10, 15, 20 or 30. 5. Click **Generate 10 ideas** (the button names the count you chose). **Advanced settings** holds three extras: a **Model** to generate with, a **View count threshold** that sets how many views a source video needs before it counts, and **Custom requirements**, up to 400 characters of free-text steering such as an audience or a tone. The channel's **Channel brief** also feeds this — see [Channels](/docs/channels). ## Results Ideas arrive under **All-time popular topics** and **Trending topics**, each with the **Example videos analyzed** that produced them. The bookmark icon on an idea saves the title for later. Every previous run for the channel is listed below the form and can be reloaded, form settings and all. ## Turn an idea into a project Click **Use** on an idea. The **Use Idea** dialog offers three routes: - **Create one-shot video** — runs the whole pipeline from this title. See [Pipeline overview](/docs/pipeline-overview). - **Create Script** — opens the script step with the title filled in. See [Script](/docs/script). - **Rewrite title with AI...** — generates variations, then pick the one you want before choosing a route. --- # Competitors & outliers URL: https://autotube.pro/docs/competitors Section: Channels & research > Track the channels you compete with, spot the videos that broke out on them, and keep a swipe file of anything worth stealing. Competitors are the YouTube channels AutoTube reads to ground its ideas. Each one is attached to a channel of yours, and the videos it publishes feed ideation and outlier detection. ## Add competitors Open **Competitors** from the **Channels** menu, or the **Competitors** section on a channel's own page. Click **Add competitor**, choose the channel to attach it to, and paste the YouTube channel URL. **Add Another Competitor** adds more in the same pass. You can also add one while creating a channel — see [Channels](/docs/channels). The list shows each competitor's **Subs**, **Videos**, **Views** and **Avg** views per video, when the channel **Joined** YouTube and when you **Added** it. Every column sorts. Pick a channel of yours from the pills at the top, or **All channels** to see everything at once. The trash icon on a row removes a competitor from that channel. ## Outliers An outlier is a recent competitor video that has clearly outrun the average video on its own channel — a breakout worth reading as a signal rather than a one-off. AutoTube detects them from the view counts it already caches, with no work from you. Open **Ideation** and switch to **Competitor outliers**. Pick one of your channels or **All channels**; outliers are grouped by the channel they belong to. Click **Use** on one to open the **Use Idea** dialog and turn it into a script or a one-shot video, exactly as in [Ideation](/docs/ideation). If the list is empty, you have no competitors saved yet, or none of them has broken out recently. ## Swipefile **Swipefile**, in the account menu, is the manual counterpart: videos and channels you saved because they were interesting, not because a score found them. Switch between **Videos** and **Channels**, then click **Add video** or **Add channel**. Paste the YouTube URL, optionally attach it to one of your channels, and add **Notes**. From a saved row you can edit the notes, remove it, or promote it with **Add as competitor** so it starts feeding ideation. --- # Thumbnails URL: https://autotube.pro/docs/thumbnails Section: Tools > Generate thumbnail ideas and images with AI, or compose one by hand in the canvas editor, then save it to a project. Two screens make thumbnails, both under **Tools**. **Thumbnail Generator** writes ideas and generates images from them. **Thumbnail Designer** is a canvas editor for laying one out yourself. Either can save its result as a project's thumbnail. ## Thumbnail Generator The header switches between two modes. **Generate from Project** starts from a script you already have. Pick a **Project**, set **Number of ideas** (3, 5, 10 or 15), and optionally add **Requirements (optional)** — anything the ideas should respect. Two extras shape the look: **Include character from project** uses the project's character as an image reference, and **Upload or Paste Images** takes up to two style references. Tick **Auto-generate thumbnails when ideas are ready** to skip a step. Click **Generate ideas**. Each idea appears with its own **Generate Image** button, or **Generate All** does the whole batch. Every generate button shows its credit cost before you click it. **Generate from YouTube Video** starts from a thumbnail that already exists. Paste a YouTube URL under **Thumbnail source** and pick a **Mode**: **Recreate** analyses the thumbnail and regenerates something similar, **Edit** modifies the image directly. Recreate runs **Analyze thumbnail** first, leaves the **Thumbnail analysis** editable, then generates from it with **Generate thumbnail**. Edit takes instructions and **Edit thumbnail**. ## Thumbnail Designer The designer is a canvas, one tab per thumbnail, kept in your browser. **Add Elements** gives you **Add Text**, **Add Image**, **Add Rectangle**, **Add Circle** and arrow shapes. The **Elements** panel is the layer list — hide, duplicate, reorder or delete. **Properties** edits whatever is selected: font, colour, gradient, size, crop, background. Undo and redo are on the toolbar. Export the canvas at **1080p** or **720p** to download it. Images generated in the Thumbnail Generator have an **Open in Canvas** action, so you can finish an AI image by hand. ## Save it to a project In the generator, the save icon on a generated image is **Save as Thumbnail for project**. In the designer, **Save as Project Thumbnail** opens a dialog where you filter by channel, **Select Project** and click **Save thumbnail**. The thumbnail then shows on the project — see [Projects](/docs/projects) and [Description & publish](/docs/description). --- # Video Editor URL: https://autotube.pro/docs/video-editor Section: Tools > Open a project's visuals and voiceover on a timeline, adjust the cut, add overlays and captions, then render again. The video editor is a timeline over work the pipeline already produced. It does not replace the [Render](/docs/render) step — it gives you a place to change the cut before rendering it again. Open the [Video Editor](/app/video-editor). The toolbar has three pickers — **Project**, **Media Run** and **Voiceover** — which decide what the timeline is built from. Changing the media run or the voiceover builds a separate set of edits, so switching back and forth loses nothing. ## The timeline Rows run down the left: **Main Video Track** for the scene clips, **Voiceover**, a music track, and any overlay tracks you add (up to three). Above them sits a ruler with chapter markers and a strip of beats taken from the voiceover, so you can line a cut up with a sentence. - Drag a clip to move it; drag its edge to trim it. - Drag the playhead to scrub; the transport buttons play, pause and skip to start or end. - Zoom with the slider, or **Fit full timeline**. - Each track header adds things at the playhead: **Add graphic at playhead**, **Add overlay at playhead**, **Add text overlay at playhead**. Select a clip or overlay and the inspector on the right edits just that one. ## Media, subtitles and style The left rail holds four sections. **Media** is the project's asset bin — drag an asset onto the timeline, or **Add media** to upload your own image or video. **Subtitles** picks a **Style preset** and can write captions from the voiceover with **Generate AI captions** at **Minimal**, **Normal** or **Dense** density. **Project style** sets a **Style preset** for the whole timeline. **Green Screen** turns on composite mode, which replaces the main track with a foreground and a B-roll track. Templates are the motion graphics and clip effects you add from the timeline; you pick one when you add a graphic. ## Render and reset Your edits save by themselves — the toolbar shows **Saving…** and then **Saved**, and there is no save button. When the cut is right, click **Render**. The **Confirm Render** dialog lists the title, duration, resolution, frame rate and the estimated credit cost before you commit; see [Credits & usage](/docs/credits-and-usage). When it finishes, **Your video is ready** offers **Open in new tab** and **Download**. The gear menu has **Clear all overlays** and **Reset timeline…**, which throws away every edit for the current project, voiceover and visuals and starts again from the sources. --- # AI Playground URL: https://autotube.pro/docs/ai-playground Section: Tools > Generate images, video clips and music from a prompt, outside any project, to test a model or a look before you commit. **AI Playground**, under **Tools**, is a scratchpad for the generation models. Nothing here belongs to a project or a [channel](/docs/channels) — it is where you try a prompt, compare models and see what a setting does before using it on a real video. ## The three modes The segmented control in the header switches between **Images**, **Videos** and **Music**. All three run at once: a generation started in one mode keeps going while you work in another, and every finished run stays in the feed above the composer. Type into the composer at the bottom of the page and press Enter, or click the generate button. The options row inside the composer changes with the model you pick — aspect ratio, resolution, duration, frames per second, number of images, and the model's own settings. ## Images Describe the image and generate. Attach reference images by dragging them in or pasting, where the selected model supports them. Switching the prompt field to multi-prompt mode treats **one prompt per line** and generates one image for each. Click a result to open it full size and step through every image with the arrow keys. Generated images can be edited in place; the original is kept, so you can toggle back to it. ## Videos Describe the shot, or the camera movement if you are animating a still. Some video models need a source image rather than a prompt alone; the composer says so before the generate button goes dead, so you always know what is missing. ## Music Describe the mood, instruments and tags in the main field. The **Lyrics** field below takes either real lyrics or `[instrumental]`. ## Cost Generating images, video and music uses credits. See [Credits & usage](/docs/credits-and-usage), or [Bring your own API keys](/docs/byok) if you want these calls to run on your own provider account instead. --- # Projects URL: https://autotube.pro/docs/projects Section: Projects > A project is one video's workspace — its script, voiceover, media, renders, thumbnail and description in one place. Everything a single video needs lives in a project: the script, the voiceover, the generated media, every render made from them, the thumbnail and the YouTube description. **Projects** is the list of them. ## The list Each row shows the project **Name**, its **Channel**, the script **Length** and when it was **Created**, with a thumbnail if it has one. Icons on the row link straight to a project's voiceover, media and render, and to the YouTube video once it is published. Filter by clicking a channel pill above the list, or search by title. Projects with work currently running appear in an **In Progress** section above the table, and drop back into it when they finish. **Load more** fetches the next page. ## Create a project Click **New project** and pick one of two routes: - **Generate script with AI** — write a new script from a title. See [Script](/docs/script). - **Import your own script** — paste a script you already have, and upload your own voiceover with it. See [Use your own script or voiceover](/docs/use-your-own). An idea from [Ideation](/docs/ideation) creates a project too, either as a script or as a one-shot run of the whole pipeline. ## The assets dialog The eye icon, **View project**, opens everything the project holds without leaving the list: - The script, with **Copy** and **Edit**, and **Load full script** for long ones. - The thumbnail, uploadable or replaceable in place. See [Thumbnails](/docs/thumbnails). - Every **render**, with playback and download. - Thumbnail propositions, generated media runs and voiceover tasks. - The YouTube description and tags. See [Description & publish](/docs/description). The menu beside the title adds **Rename**, **Show metadata**, **Show chapters** and, once the video is up, **Open YouTube video**. **Edit project** opens the full-page script editor instead, where you can change the title, the script body and the channel the project belongs to. ## Voiceovers **View voiceover jobs**, on the [voiceover step](/docs/voiceover), opens a cross-project list of every voiceover job with its status, project and channel. Filter it by channel or by status, or search it. ## Delete The row menu deletes a project. The confirmation lists what goes with it under **Also permanently removed** — renders, media generation runs, voiceovers and the custom thumbnail — and deletion cannot be undone, so download anything you still want first. A project with work still running cannot be deleted until it finishes. --- # YouTube URL: https://autotube.pro/docs/youtube Section: Integrations > Link a channel to the real YouTube channel it publishes to, so ideas and briefs have real numbers to work from. An AutoTube [channel](/docs/channels) is a set of defaults with a name you chose. Linking it to the YouTube channel it actually publishes to gives the app real numbers to reason about — your own titles, view counts and thumbnails — rather than guesses. ## Connect a channel 1. Open **Channels** and click **Settings** on the channel, or open the channel and click **Settings**. 2. On the **General** tab, fill in **YouTube Channel ID or URL**. Either form works: a channel ID like `UC_x5XG1OV2P6uZZ5FSM9Ttw`, or a full URL such as `https://www.youtube.com/@YourChannel`. 3. Click **Save settings**. Once it resolves, the channel's page shows the real channel under its title — picture, name, handle and subscriber count — and the line links straight to YouTube. That line is how you confirm you linked the right one. **YouTube Playlist ID** is a separate, optional field for pointing the channel at one playlist. ## What it enables - **Ideation.** Your own channel joins the **Sources** list alongside the competitors you track, so ideas can be grounded in what already worked for you. See [Ideation](/docs/ideation). - **Channel brief.** **Generate** on the **Channel brief** field reads the connected channel's top-performing titles to write the brief. Without a connection it falls back to your recent AutoTube scripts, and needs a minimum number of either before it will run. - **Your AI assistant.** The MCP tools read the connected channel's YouTube data directly, which is what makes an assistant's suggestions specific rather than generic. See [MCP server & AI skills](/docs/mcp-and-skills). ## What it does not do Linking is read-only. The app never posts to your channel: you download the finished render and upload it yourself, as described in [Description & publish](/docs/description). No YouTube sign-in or permission grant is involved, so there is nothing to revoke on Google's side. ## Disconnect Clear the **YouTube Channel ID or URL** field and save. Nothing else is deleted: the channel, its projects and its competitors stay exactly as they were, and the features above fall back to what they do without a connection. --- # Bring your own API keys URL: https://autotube.pro/docs/byok Section: Integrations > Connect your own provider accounts so the steps they cover cost no credits and you pay the provider directly. AutoTube normally runs every generation on its own provider accounts and bills you in credits. Bring your own key and a step that your key covers costs zero credits — the provider charges you instead. This is what the **Own Keys** plan is for; see [Plans & billing](/docs/plans-and-billing). ## Where keys go Open **Account & Settings** from the account menu and select the **API Keys** tab. Each provider has a row. | Provider | What your key covers | | --- | --- | | Google AI | Nano Banana image models and Gemini image-prompt generation | | OpenAI | Script generation on GPT models | | Anthropic | Script generation on Claude models | | xAI | Grok outline and script generation, Grok Imagine images and image prompts | | Replicate | Replicate image and video generations, plus Qwen3 and Chatterbox voiceover | | ElevenLabs | The ElevenLabs voiceover provider | Click **+ Add key**, paste the key and save. Keys are stored encrypted. Once a key is saved the row offers **Change**, **Test** — which calls the provider to check the key works — and **Remove**. There is no on/off switch: a saved key is used, a removed key is not. Each row also links out to the page where that provider issues keys. ## What still costs credits Bringing your own keys changes who pays for a model call, not what the app does. Credits are still spent on: - Any model you have not brought a key for, at the normal rate. - Stock-footage picking, on every plan. - Rendering past your plan's daily included render allowance, and skip-the-line rendering at any time. A plan that is not **Own Keys** hides providers it does not entitle. If you already stored a key for one of those, the row stays visible, reads **Not used on this plan**, and offers only **Remove**. ## Usage history Below the keys, **Recent usage** lists the last generations that ran on your keys, each marked **Succeeded** or **Failed**. The link opens the full **API key usage history**. The same log has its own page, [API key usage](/app/byok-history). It lists every generation that ran on your own keys with the **API**, **When**, **Model**, **Tokens**, **Status**, **Project**, **Media run** and **Error** for each, filterable by provider and sortable by API, time or status. Use it to reconcile a provider's bill against what AutoTube actually ran. For what credits themselves pay for, see [Credits & usage](/docs/credits-and-usage). --- # MCP server & AI skills URL: https://autotube.pro/docs/mcp-and-skills Section: Integrations > Connect Claude or another assistant to AutoTube over MCP, and install the skills that teach it how to run the workflows. AutoTube runs an MCP server. Connect an AI assistant to it and the assistant can read your channels, projects, competitors and presets, and start real work — scripts, image prompts, whole videos — without you leaving the chat. Two halves make it useful. **MCP** is the connection that gives an assistant the tools. **Skills** are instruction files that teach it how to use them well. Skills alone still read; only MCP lets them act. ## Connect an assistant Open **Account & Settings** from the account menu and select the **MCP** tab. There are two ways in. **Claude custom connector** needs nothing but the URL shown under **Remote MCP server URL**: 1. In Claude, open Settings → Connectors → Add custom connector. 2. Paste the URL and add it. 3. Sign in to AutoTube when prompted. No client ID or secret is needed. Connected clients are listed below with what they can reach — **Read-only** or **Read + video write** — when they connected and when they were last used. The trash icon revokes one immediately. **MCP token** is for Codex and any other client that does not do OAuth. Click **Create token**, copy the bearer token, and paste it into that client's configuration. The token is shown once; delete it and create a new one to rotate access. ## The four skills **Resources**, under **Tools**, lists the skills. Each card has a **Copy prompt** button — paste that prompt into your assistant and it fetches and installs the skill itself. | Skill | What it does | | --- | --- | | Long-Form Script Writer | Outlines first, writes section by section carrying continuity forward, researches facts, then polishes | | YouTube Ideation | Ranks video ideas from channel fit, recent winners and losers, competitor gaps and work already in progress | | Image Prompt Writer | Writes every scene's image prompt for a project, then imports them | | AutoTube Full Video Creation | Writes or imports a script, then drives voiceover, visuals, render and description | Two of them have prerequisites. Image Prompt Writer needs a project whose voiceover is finished, because AutoTube splits the scenes from the voiceover. Full video creation runs on the channel's saved default presets — Script, Voiceover, Visuals, Render and Description — and fails if any is unset. Set them on the **Default Presets** tab described in [Channels](/docs/channels). Connect the channel to its real YouTube channel too: the assistant then reads YouTube data directly and its suggestions get sharper. See [YouTube](/docs/youtube). Work an assistant starts spends your credits exactly as the app does. See [Credits & usage](/docs/credits-and-usage). --- # Plans & billing URL: https://autotube.pro/docs/plans-and-billing Section: Account > Pick a plan, switch it, buy credits and manage payment — all of it settled through Stripe. A subscription is what gives you access to the app; credits are what pay for each generation. This page covers the subscription. For what credits buy, see [Credits & usage](/docs/credits-and-usage). ## The plans Three plans are on offer: **Creator**, **Pro** and **Own Keys**. Each is billed **Monthly** or **Annual**, and annual comes with months free — except Own Keys while its early-bird monthly price runs, which is cheaper than its annual price. Every plan includes the full pipeline, any video length, premium voices and voice clones, multiple channels from one dashboard, credits that carry over, and Discord and email support. What differs is the ceilings — channels, visuals per video, script length, stock clips — and the monthly credit grant. **Own Keys** is the specialist option: you connect your own provider accounts and pay them directly for every generation those keys cover. It suits people who already run those accounts. If you are unsure, take Pro — the same plan with everything included. See [Bring your own API keys](/docs/byok). Current prices are on [the pricing page](/pricing). ## Subscribe or change plan **Plans & Billing** in the account menu opens the billing screen. Under **Plans**, the card for your current plan reads **Current plan**; the others offer **Get Creator** or **Switch to Pro**, depending on whether you have a live subscription. Switching opens the Stripe portal with the change ready to confirm. A switch to a plan with a lower list price (before any discount), or from annual to monthly, takes effect at the end of your current period — the billing screen then shows **Switches to** and the date. Any other switch applies right away, prorated for the rest of the period, and a bigger credit grant starts with your next grant. Credits you already hold are never touched by a plan change. Annual is billed once, but credits still arrive one month at a time rather than twelve up front. The **Next credit grant** row tells you when the next one lands. ## Buy credits Click **Buy credits** on the billing screen, or **Buy Credits** in the account menu. Pick a package or enter a custom amount, then **Continue**. Credits bought this way are a one-time purchase and roll over while your subscription is active. The larger packages carry a bonus, shown on the tile. ## Payment, invoices and cancelling **Manage subscription on Stripe** opens the Stripe customer portal, which is where payment methods, billing details, invoices and cancellation live. If a payment fails, the billing screen says so and the button becomes **Update payment method**; you can still view existing work, but new creation and credit spending pause until billing is fixed. Cancelling is done in the same portal. The subscription stays active until the end of the paid period, and credits already granted stay on your account for 30 days — they are usable again if you resubscribe in that time; after that, unused credits expire. Refunds cover unused credits only; anything already spent on a generation is not refundable. --- # Credits & usage URL: https://autotube.pro/docs/credits-and-usage Section: Account > Credits pay for every generation; this is what they buy, how reservations work and where to see what you spent. Credits are the unit every generation is billed in. Your balance sits in the top bar, and the app shows the estimated cost before anything expensive starts. ## What credits pay for Every AI step bills credits: script outlines and scripts, image prompts, image, video and music generation, visuals runs, voiceovers and voiceover transcription, video descriptions, custom graphics, YouTube transcripts and video renders. Buying credits and any refund or adjustment show up in the same history. Unused credits roll over from one billing period to the next while your subscription is active. If your subscription ends, your unused credits are kept for 30 days and are all still there if you resubscribe in that time; after that they expire. Where they come from — a plan's monthly grant or a one-time purchase — is covered in [Plans & billing](/docs/plans-and-billing). ## Reservations When a long task starts, AutoTube sets aside an estimate of what it will cost. That set-aside is a **Reservation**, and it behaves differently from a charge: - It does not change the balance you see in the app. - The real cost is deducted when the task finishes. - A task that fails releases its reservation. A failed generation is not charged. You will meet reservations in two places. They appear as rows labelled **Reservation** in your credit history. And if you run low while work is still running, the error says so exactly — how many credits the new task needs, how many you hold, and how many are reserved for tasks still going. Wait for those to finish and the reserved credits come back. ## Where to see your usage Open **Buy Credits** from the account menu and click **History**. **Credit history** has two views: **History**, a row per transaction with its type and amount, and **Per project**, which groups spending by the video it went on. **Load more** pages back through it. If you run generations on your own provider keys, those have their own log — see [Bring your own API keys](/docs/byok). ## Daily included render On the plans that include it, part of your rendering is covered by a daily allowance rather than by credits. The billing screen shows it under **Included render**: how much you have used today, and when it resets. Rendering at 1080p draws on the allowance at twice the rate of 720p. Before a render starts, the confirmation dialog says which case you are in — **Included in your plan**, a partial cover where the remainder bills credits, or **Daily included render used up**. Skip-the-line rendering always bills credits, allowance or not. See [Render](/docs/render). --- # Collaborators URL: https://autotube.pro/docs/collaborators Section: Account > Share a channel with someone by email, choose how much they can do, and remove them when you are finished. A channel can be shared with other people. They sign in with their own Google account, see only the channels you shared, and work inside them at the level you chose. Collaborators do not need a subscription of their own. ## Invite someone Open the channel and scroll to the **Members** section — it is on the channel's own page, and only its owner sees it. Enter their **Email**, pick a **Role**, and click **Invite**. **Also invite to** lets you add them to your other channels in the same action. The email must be the Google address they will sign in with. They get a message inviting them to AutoTube; the invite stays **pending** until they sign in with that address. If they are already signed in to AutoTube with it, they need to sign out and back in before the shared channel appears. Your plan includes a number of collaborators across all your channels, pending invites included. The **Members** section states the figure. ## Roles | Role | Can do | | --- | --- | | Viewer | Open projects and download renders and assets | | Assistant | Everything a Viewer can, plus update thumbnails, titles, descriptions and tags | | Editor | Everything an Assistant can, plus create and produce videos | An Editor spends the channel owner's credits — see [Credits & usage](/docs/credits-and-usage). Some things stay with the owner whatever role you give: deleting projects, managing competitors, changing channel settings and presets, managing members, billing and API keys. See [Channels](/docs/channels). ## Manage and remove Each member row carries a role dropdown — change it and the new permissions apply at once. The button at the end of the row is **Revoke** for an invite that has not been accepted, and **Remove** for someone who has joined. Removing takes effect immediately. ## What a collaborator sees A shared channel appears in their **Channels** list marked **Shared by** the owner's name. They can reach its projects and create work in it, within their role. They cannot see your credits, your billing or your other channels. **Leave channel** on the shared channel's page gives up access themselves. See [Channels](/docs/channels) for everything else a channel holds. --- # Support & community URL: https://autotube.pro/docs/support Section: Help > Where to ask a question, report a bug, and find the in-app guide and video tutorials. Something broken, or stuck on a step? There are three places to look, and one of them is inside the app. ## Get help The question-mark button in the top bar opens the in-app help dialog. **Quick Start** walks the six steps from creating a channel to making a thumbnail; **Video Tutorials** holds walkthroughs, including building a first video and case studies of real channels. Beyond that: - **Discord** — the fastest route, and where other creators answer too. The community is for subscribers; the Discord icon in the top bar of the app takes you to the invite. - **Email** — [support@autotube.pro](mailto:support@autotube.pro), also linked at the foot of the help dialog. ## Reporting a bug Include enough for someone to find the run: - The link to the project, or its title and channel. - Roughly when it happened, and your timezone. - What you clicked and what happened instead. - The exact error text, if there was one. A project link beats a screenshot: it identifies the exact script, voiceover, media run and render behind the problem. See [Projects](/docs/projects). ## For AI assistants These docs are published in a machine-readable form as well — see [For AI assistants](/docs/ai-access). --- # For AI assistants URL: https://autotube.pro/docs/ai-access Section: Help > The docs in machine-readable form, so an assistant or crawler answers autotube.pro questions from the same text you read here. These docs are also published as plain Markdown files, following the [llms.txt](https://llmstxt.org) convention. Point an assistant at them and it answers from what the app actually does, not from a guess. ## The files - [llms.txt](/llms.txt) — an index of this documentation: one line per page with its URL and summary. - [llms-full.txt](/llms-full.txt) — the complete documentation in a single Markdown file, for loading into an assistant's knowledge base. Both are generated from these pages on every request, so they never lag behind what you are reading. ## When to use which - Building a support bot or a custom GPT: ingest **llms-full.txt**. It holds every page, table and option in one document. - Letting an agent browse: give it **llms.txt** and let it fetch the pages it needs. - Working in Claude Code, Cursor or another coding agent: the [MCP server & AI skills](/docs/mcp-and-skills) page is the one you want. It connects the assistant to your account, not just to the docs. ## Questions the files cannot answer Anything about your account, your channels or a specific project lives behind sign-in and is not in these files. Ask in the app, or through the MCP server, which reads your data with your permission. For everything else, see [Support & community](/docs/support).