> For the complete documentation index, see [llms.txt](https://thecontentforge.gitbook.io/thecontentforge-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://thecontentforge.gitbook.io/thecontentforge-docs/feature-guides/video-studio.md). # Video Forge Video Forge turns an uploaded interview, podcast, webinar, or clip into platform-ready content. It handles the media pipeline end-to-end: upload, processing, transcription, transcript analysis, visual enrichment, built-in clipping, and multi-platform post generation calibrated to your Brand Voice. It also includes **Create**, a full short-form video editor for turning the best moments into captioned, branded shorts you can export as MP4 - see [Video Forge - Create](/thecontentforge-docs/feature-guides/video-studio/video-forge-create.md). ### The pipeline Every upload moves through five stages. You can watch progress live on the session page; the tracker auto-advances as each stage lands. | Stage | What happens | | -------------- | ------------------------------------------------------------------------------------------------------- | | **Uploading** | Your file uploads securely, with progress shown live | | **Processing** | The video is prepared for smooth playback | | **Captioning** | Subtitles are generated; the transcript is fetched, stored, and made available in the session workspace | | **Analyzing** | AI extracts summary, key topics, hooks, strongest quotes, key moments, and follow-up angles | | **Ready** | The session can generate content and visual insights | Long videos are handled reliably end-to-end - a 90-minute interview analyzes completely without timing out. ### Session workspace Opening a session gives you a player, transcript, and workspace tabs: * **Player + transcript** - searchable transcript with clickable timestamps and copy controls * **Insights tab** - summary, key topics, strongest quotes with timestamps, best hooks, key moments, and follow-up angles * **Visual tab** - visual summary, alt text, visual chapters, on-screen text, chart or screen references, and clip suggestions when the live video asset is still available * **Content tab** - the video overview and generated posts; returning to a session with generated posts opens this tab automatically * **Generate tab** - multi-select menu of output types; only output types that make sense for the video's duration are shown * **Create tab** - the built-in short-form video editor (badged **NEW**). Cut a moment into a captioned, branded vertical, square, or landscape short and export it as an MP4. Appears once the upload is processed and playable. See [Video Forge - Create](/thecontentforge-docs/feature-guides/video-studio/video-forge-create.md) Visual insights require the live video asset. If the source video has already been pruned, transcript and written insights remain available, but visual enrichment and thumbnail remixing cannot use the original frames. ### Built-in clipping and thumbnails > **Quick clip vs. the Create editor:** the Visual tab's **Create clip** below is the fast way to cut a raw MP4 from a single range. For full short-form editing - captions, B-roll, music, layouts, effects, and a styled render - use the [Create](/thecontentforge-docs/feature-guides/video-studio/video-forge-create.md) tab instead. The Visual tab can suggest clippable moments. Use **Create clip** to cut a short MP4 from a suggested range, or **Use in X post** to jump straight to a clip-scoped X post. Every cut clip then appears in the **Clips** rail at the top of the Content tab, where a clip is a first-class content source: * **Generate from clip** - opens the Generate tab scoped to the clip. Outputs are grounded in the moment you cut, so the post is about the clip rather than the whole recording, and the output menu keys off the clip's length, not the full video's. Clip-scoped posts show a scissors badge with the window. * **Auto-caption** - renders a captioned MP4 of the clip (pick 9:16, 1:1, 16:9, or 4:5) using the Create pipeline's default caption style, with no editor visit. Word timings are generated automatically if the window doesn't have them yet, and when the output shape differs from the source (e.g. a 16:9 interview cut to 9:16) the face-aware auto-reframe runs first so the crop follows the speaker instead of staying locked to the center. The result downloads right from the rail; **Re-run** renders again (e.g. after changing the aspect), and the underlying project stays editable in the Create tab if you want to restyle it. * **Use in X post** - Generate tab preset with only X Post ticked, scoped to the clip. * **Download MP4** - the raw cut, ready to save. * **Save to library** - imports the MP4 into the [Asset Library](https://github.com/fecktk1/social-os/tree/main/docs/features/brand-assets.md) so it can be attached to any post and survives source-video retention pruning. When you publish a clip-scoped post, the clip MP4 is imported and pre-attached automatically - no downloading and re-uploading. You can remove or swap the attachment in the publish panel like any other media. Generated content cards can also open **Remix Thumbnail**. The remix tool starts from the best available frame of your video and lets you refine or download the result. ### Fast vs Best Same toggle as [Content Forge](/thecontentforge-docs/feature-guides/content-studio.md): **Fast** is the quickest, **Best** takes longer for higher quality. Generate defaults to **Best**. The choice persists per user. ### History The **History** button at the top of a session opens a floating overlay listing every prior video session for the org. Pick one to switch sessions without losing your place; start a fresh upload from the same overlay. ### What the Generate tab writes Each selected output becomes its own row in the [Generation Queue](/thecontentforge-docs/feature-guides/generation-queue.md) and then appears in the Content tab when done. All outputs are grounded in the transcript and AI-extracted insights, then passed through your [Brand Voice](/thecontentforge-docs/feature-guides/brand-voice.md). Video Forge can generate: * **X Post** * **Quote Card** * **Key Takeaways** * **Instagram Caption** * **Facebook Post** * **Telegram Update** * **Teaser Copy** * **Clip Titles** * **Follow-Up Angles** * **X Thread** * **Episode Summary** * **Episode Title** * **Clip Hooks** * **Description Copy** * **Thumbnail Text** Short clips show the fastest-fitting options first. Some outputs only appear once the video is long enough - longer videos unlock options such as Clip Titles, X Thread, Episode Summary, Episode Title, Clip Hooks, Description Copy, and Thumbnail Text. ### Publishing Publishable outputs can be copied, revised, remixed when applicable, published now, scheduled for later, or manually logged after posting elsewhere. If your org does not use an approval workflow, any user can publish through an already-connected publishable profile. If approval is enabled, send behavior is gated by the configured minimum approver role. Scheduled, pending-approval, posted, failed, and manually logged posts appear on the [Calendar](/thecontentforge-docs/feature-guides/calendar.md). ### Deletion and retention A **Cancel** / **Delete** button on the session header stops running work and removes the transcript, insights, and drafts. If the session is attached to a pending post, cancel that post first. Transcripts are available for 60 days after upload. Source video assets are retained based on your plan or custom override; the default source-video windows are 48 hours on Starter, 120 hours on Pro, and 288 hours on Enterprise. Pruned sessions stay in history for reference, but visual features that need the live video source are unavailable after pruning. ### Monthly upload cap Each plan tier has a monthly cap on completed uploads. **Failed uploads do not count** - only sessions that reach the processed stage or beyond are billed against the cap. If an upload errors out mid-stream, retry without burning a slot. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://thecontentforge.gitbook.io/thecontentforge-docs/feature-guides/video-studio.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.