play_circleGetting started

The rail has two halves. The top is the day's work, in order: AI Home (the assistant), Gallery (your library and working board, the home screen), ContentConcierge (what you have asked us to make, and what came back), and Studio (every ContentCanvas your team is making right now). The bottom is the setup you visit once and then rarely: Automations, Integrations (connected apps), Brand and Train, followed by search, Settings, Help (which opens these guides), and your Account.

Press Cmd+K (Ctrl+K on Windows) at any time to jump to any destination, any Job, or any piece of content by name, from anywhere in the app.

Start a ContentCanvas from a source on the Gallery board, produce a single asset from the Studio, or hand the work to AI Home in plain language. All three end in the same place: real, editable content filed into your Gallery with a record of what it was built from.

Drag a file from your machine straight onto the Gallery board. It uploads, shows itself being read, and lands as a source split into sections. PDFs, documents, decks, images, audio, and video are all accepted; a file the platform cannot take says so plainly rather than failing quietly.

Point the platform at a URL to read one live page, import a whole website's content in bulk, or run a research job to have it gather and cite current material across the open web. All three arrive as ordinary sources in the same library.

Connect a third-party app on Integrations, on the left rail. Its items then appear as their own tab in your Gallery, and importing one turns it into a source split into sections, exactly like an upload.

The tabs across the top are grouped by where things came from: what you singled out (Starred, Folders), what you brought in (Source content, Research, Website), what is plugged in (each connected app), and what the platform made (Generated, Collections). Each card carries a cover, its format icon, and its stage.

Switch layouts with the toggle: Canvas is the spatial board and the default, Grid shows covers, List is dense for scanning, and Graph shows lineage so you can see which assets came from which source. On a phone you land on List.

Type in the search box to find anything by keyword or by meaning. Filter Generated work by workflow stage, or open Collections to slice the whole library by the kind of material inside it. Clicking a source on the board fans out everything made from it; clicking a production run opens what that run produced.

Tabs, layouts, the thing you drilled into, even the section you selected are all in the address bar. Back and Forward work, a refresh restores exactly what was on screen, and you can send a colleague a link to precisely what you are looking at.

Open Train and answer one question: what does this design make? That name (a Webinar promo, an Event one-pager) becomes the template's name and the name of everything it produces. Then drop in the finished piece itself: an image, a PDF of up to 20 pages (a multi-page PDF trains as a deck, one slide per page), a PowerPoint (.pptx), an InDesign IDML package, an HTML or Word document, or plain text. Or paste a design link instead of a file, and a Figma file or a Google Slides deck trains natively through your connected app. Press Train this template.

Training is a tracked run you can watch in your Activity panel, and it takes a couple of minutes. The platform reads the design, lifts every word, logo, and button off the artwork while keeping the artwork itself, names each text element the way a designer would (Headline, Subhead, Call to action), and rebuilds the whole thing as an editable template. A design that arrived natively (PowerPoint, IDML, Figma, Slides) keeps its exact positions and type sizes, because the file states them rather than a picture implying them. You can close the tab: the tile on Train shows live progress and survives a reload.

Drop in a whole PowerPoint master rather than one design and you get something different: a LAYOUT LIBRARY. Every slide becomes a pattern available for new work, grouped by the deck's own PowerPoint sections, and sections named for people rather than patterns (How to Use, Typesetting) are trained but never picked. Its card says so instead of counting fields, as in 97 slides, 12 sections, a layout library. A big master takes about ten minutes and narrates its real progress while it works.

Every template is a tile: the cleaned artwork as its cover, its name, and how many fields it has to fill. A tile still building says so, and a failed one tells you why in plain words. Open a ready template to fine-tune it in its editor: move and resize elements, restack layers, adjust type and color. The base artwork is the frame's background, so you can never grab or delete it by accident.

The moment a template is ready, a Make Job named after it appears for your whole team: in production runs, in AI Home, and in every Automations picker. Removing a template stops it being offered, nothing more: the style it learned and every asset it produced are kept.

Open Produce and pick the template, then the source it should be about, and optionally the exact material inside that source. Draft the copy and every field fills in, ready to edit, with the original copy shown beneath each one: the original is the specification for how much text that space holds. Nothing is made until you press Proceed, and then you land straight in the finished asset's editor.

An image field offers three ways in: pick from your existing content and brand assets, upload a file, or generate with AI. On a graphic, a field you leave empty leaves that element off the asset rather than rendering a blank; an HTML or text template keeps the picture it was trained with.

On a graphic template you can change the artwork itself before producing. Drag a box over the part of the picture you mean, describe the change, add as many regions as you like, then apply them in one AI pass. The edits are confined to the regions you drew; the rest of the artwork is untouched.

To produce a batch, start a production run from any source: from its card on the board, from its page, or from AI Home. The platform pulls the source's usable material, you keep all of it or pick the parts that matter, then choose one or more templates; each becomes one asset, built as a live tile under the source. The run belongs to the server, so you can close the tab. If a template you picked is still training, the run waits for it and shows the training's own progress in the tile.

Every produced asset carries a small plus beneath it on the board: describe the changes and a new version is produced, chained under its parent. And an uploaded image of a finished design can become an editable copy in one click with Create a new version of this source; the platform trains the template the first time and reuses it after that.

The asset opens in View: a clean proof of the finished work on a dark mat, at the size and shape it really is. Toggle to Edit in place. There is no separate viewer and editor to keep straight, and the preview adapts to the medium, so a deck pages as a deck, a document reads as a document, and video and audio play.

Covers are made from the asset itself, not a screenshot of a page. A LinkedIn post covers as the feed card it will be: your brand name and avatar, the copy folded where LinkedIn folds it, the creative edge to edge, and the reaction bar.

Each asset records what it was built from, so you can trace any piece of finished work back to the source, the run, and the template that made it, from the asset page or from the Graph layout.

Download the asset in the formats its type supports, take everything a run produced with Download all, or mint a share link that anyone can open with no account. Unless publishing has been switched on for your workspace, the platform does not post to your channels on your behalf: you take the finished file and publish it yourself. With it on, a publish Job can send an approved asset to a connected channel, and every post still waits for a person to say yes.

Most deliverables open into the section editor: the outline on one side, a live preview on the other, kept in step. Click something in the preview to find it in the outline and the other way round. Click a heading or paragraph and type; there is no Edit button and no overlay.

Ask AI to rewrite, shorten, or restyle one section without touching the rest. Every section keeps its own version history, so you can compare what changed and restore an earlier version.

A designed graphic opens on a spatial canvas: drag elements, resize with the handles, restack the layer list by drag (it shows the real front-to-back order), add pages, apply a style, or reflow with the layout picker. A deck pages slide by slide, each slide fitted to the pane. An image opens to crop, rotate, replace, regenerate, or refine a selected region with AI, archiving each version as it goes.

A video opens on a timeline to trim a range, grab a frame at the playhead as a new image, or export a section. Voiceovers and music open in the audio editor.

The Brand Voice tab holds your positioning, your tone, the exact wording you use for the things you name, and the phrases that must never appear. Every job that writes anything reads it first.

Brand Assets holds your colors, fonts, logos, and reusable artwork, with bulk upload and an import that reads a PDF of brand guidelines. If the platform has never been given a logo, it will never invent one.

Build one persona per buyer type in your committee, and list the products or services you sell. Attach a persona and an offering to a run and the asset is written for that buyer, about that offering, instead of being generically about your company.

Under Brand, Training, Examples, drop finished files in. You do not have to tell the platform what they are first: it looks at each one and works out what it is, and you only correct what it got wrong. One drop can teach several formats at once, and a logo dropped in is recognized as a logo and routed to your Brand Kit.

Each file is read by four separate passes, and a pass only runs where it has something to say. Open an example to see the findings: the file itself beside the individual elements the platform picked out of it, and what it concluded from each. You can see the evidence, not just the verdict.

Brand, Training, Styles is the Style Studio: everything Training has distilled, read back to you. The specimen at the top is painted in your own captured colors with your own voice quoted in it. Each learned rule is a card carrying its source, so you can see which example taught it, and what changed. Tokens it has never seen show as empty, never as a color you do not have.

Tap a swatch on the specimen to correct that one token. Run a test to see something made to your style right now, side by side with the example that taught it. The coverage strip shows one chip per format, so you can see at a glance what it has learned and what it has never been shown; a rule can be scoped as narrowly as one offering in one format.

An asset that has never been measured shows an invitation to check it, on the asset page and in the editor. The result is scored on four axes (color, type, logo, copy) rather than as a single number, so a low score tells you where the problem is, and it says what it measured: the whole asset, or just the part you last changed. Color covers legibility as well as palette, so text nobody can read against the surface behind it is a finding even when every color on the page is yours.

A carousel answers to all four axes. An image is held to its colors and its words, because a picture has no selectable typeface and a mark painted into it cannot be proven to be your file; a text post is a voice check. The report says which axes apply to what you are looking at and which do not, so you are never sent to fix something that was never being measured. The check is observational either way: it will never stop you approving or downloading anything, and a score from an older version of the check is labeled out of date with an offer to run it again rather than presented as current.

Send for review from the asset page, ContentConcierge, or the board. You name the people who have to sign off, and the asset waits on each of them until they respond. Reviewers outside your organization get an emailed link and can approve or request changes without an account.

An asset's status is a control you set, not a label you read. Changing it is how work moves from draft to finished. A Changes requested card says what was actually requested, not just that something was.

Every asset keeps its approval history: who was asked, who responded, what they said, and the latest change request pinned to the top. A reviewer whose invitation failed to send is shown as failed with a retry, rather than sitting there as though they are still thinking about it.

Any Make Job can run on a schedule: pick the content it should draw from, optionally give it an angle, and each fire produces one finished asset in your trained design. Run it once at a set time or on a cadence; the calendar is the default view, and any finished run can be turned into a schedule with Repeat.

On a list works down a queue, one item per run. On trigger fires when something happens, and the useful one is a new source landing: when new content arrives, make my trained design from it. Both file their output into your Gallery like any other run: a schedule produces assets, and it never posts them anywhere unless publishing is switched on for your workspace and the Job you scheduled and approved is a publish Job.

The Ideas view is your planned backlog, where content ideas you kept from an ideation job wait to be produced. History keeps a record of every execution, so you can see what ran, when, and what it made.

Brand, Integrations, Train, and Produce are rail destinations, not Settings tabs, so most members rarely need Settings at all. Organization admins manage Users, Models, and Billing there, and some workspaces also carry Usage and Performance.

Help on the rail opens the DesignTech AI guides: this getting-started guide, the full catalog of what the platform can do, a page for every integration, our privacy and security documentation, and a contact form. If the platform is down or in maintenance you will see a notice in the app itself, and the system status page is linked from Contact.

The unit of production. On Train you hand the platform one finished design that already cleared brand review, and it learns the style, the layout, and every fillable field from that file. Your workspace can only produce what it has trained: the platform never invents a look of its own, so everything it makes looks like something your team made.

What you actually run. Training a template automatically mints its Make Job, one marketing outcome with the design already decided. Alongside those, the platform ships Jobs that plan, research, and measure. Pick one, fill in a few details, and run it.

Underneath, a Job is a sequence of single operations: read the source, research the topic, draft the words, fill your template, file the result. You never run those one at a time. The Job runs them for you, in the right order. What it can do, further down, is the full inventory of the operations it has to work with.

Start a ContentCanvas from a source on the Gallery board, produce a single asset from the Studio, or hand the work to AI Home. Everything produced is filed straight into your Gallery as real, editable content, with a record of what it was built from.

Nothing is generated in the abstract. Your brand rules, your buyer committee, your offerings, and everything Training has learned from your own work are injected into every step, and the finished asset is measured back against them by the brand check.

Read one to five documents, pages, pasted passages, transcripts, or Gallery items as separate source records. Each source keeps its own identity and citation trail for the extraction steps that follow.

Set a fresh round of research beside the last one and list what is new, what changed and what went quiet, so a recurring watch reports movement rather than repeating itself.

Research an objective across the open web and return one grounded briefing of around 1,500 words, with its sources. Every claim traces back to a page it actually read; it never invents one.

Read an article, public social post, video page, app listing, or other public URL as structured evidence and file it as a source in the Gallery.

Read one live website and file its full-page screenshot, logo, public claims, and linked social pages as grounded source material.

Read one public landing page into visible copy, hero images, metadata, links, and sitelink candidates after honoring its robots.txt rules, then file that grounded source in your Gallery.

Read a public LinkedIn company page and its recent posts, rank them by engagement, and file the result into your Gallery as a source a later step can ground on.

Read a public LinkedIn person profile, its headline, about, and experience, and file it as a source document ready to build a persona from.

Turn an uploaded file into an editable asset that holds the same content and stays linked to the original. An uploaded HTML page keeps its own styles and scripts, so it opens running, exactly as it arrived.

File an approved asset as reusable source material future work can be grounded in. The asset itself is unchanged.

Find the strongest existing content in your library for a topic, by meaning as well as by keyword, and hand back a ranked list. Read-only: it never changes the library.

Turn a video or audio recording into a transcript with a word-level, timecoded timeline. Those timecodes are what transcript-driven cutting runs on.

Pull performance data from a connected ad account as structured, time-series records: impressions, clicks, CTR, spend, conversions, leads, and cost per lead, one row per campaign.

Mine a source for reusable raw material: quotes, statistics, testimonials, key insights, and headline lines, each a typed item grounded in the source rather than invented.

Read a timestamped transcript, propose grounded clip moments, and pause for you to approve the exact spans before anything is rendered.

Turn one to five sources into a structured ExtractionPlan of section arguments, key claims, every quantitative fact, attributed quotes, and a representative voice sample. Every item carries a confidence score and an exact source sentence.

Turn one landing-page source into the strict facts and language a campaign must preserve: its offer, funnel stage, promise, proof, call to action, price tier, voice, product spellings, and vocabulary rules.

Classify an uploaded design file to determine whether it is a logo or which deliverable type it represents.

Read a request together with an app’s saved settings and fill in what the run needs, or ask the one question that is still open before it starts.

Read performance data and return a plain-language read of it plus ranked, actionable recommendations, each with its rationale and a confidence.

Turn each finding in a proof report into a screenshot of the exact place in the delivered PDF, cropped to the element and outlined for review.

Compare what your Gallery library holds against what you have actually posted on your LinkedIn company page, then file a gap analysis with a ranked list of repurposes to post next.

Look at a trained template's base artwork, the design with every logo, word, and button stripped off it, and write what it is: the colors, the treatment, the shapes, the lighting, and where the empty space sits. Use it to brief a replacement background.

Continue a run only when a plain-language condition holds for the content wired into it. When it holds, the content passes through untouched; when it does not, the run stops there and says why.

Turn a request typed in plain words into a brief a producer can start on: a short title for it, the ask restated the way the producer will read it, and the questions still worth settling first, the channel, the audience, the one message it has to land, what it links to, and when it is needed. It never invents an answer; anything it does not know, it asks. Nothing is filed and no work is started, so what comes back is a brief to confirm.

Turn an extracted argument into an editable plan for sections, charts, testimonials, and imagery, then file only the treatment you approve.

Turn sourced search terms into stable ad groups by search intent, assign match types, check that every group is coherent, and file the complete campaign data in your Gallery.

Read one source and propose several content-idea briefs, each with a title, an angle, a target persona, and a rationale, then pause for you to keep the ones worth making. The kept ones land in the Planned backlog on Automations.

Given content you already own and the buyer it should target, propose a ranked set of repurposing ideas, each with a detailed rationale and the Job that produces it. Grounded strictly in what exists.

Turn a monthly budget and an average deal value into a full-funnel media mix: a realistic B2B channel split with its benchmark assumptions, then the computed funnel math (spend, impressions, clicks, leads, opportunities, revenue, cost per lead, and ROAS), filed as an editable table.

Produce a ready-to-implement LinkedIn ad setup: the objective, an audience built from real LinkedIn targeting facets and grounded in your buyer personas, the budget, the schedule, the creative, the tracking, and the exact Campaign Manager click-path. It recommends; you build it.

Inject your Brand Kit, the voice, your preferred wording and banned phrases, and the colors, fonts, and logo, as a directive every writing and image step follows. This is wiring rather than a step with an outcome of its own.

Inject the buyer persona a run is written for, so the words speak to that buyer's pains and language. Pick several and the run fans out, one asset per persona.

Extract actionable tone, grammar, and typography rules from past marketing materials to form a consistent house style.

Read a company website into a starting Brand Kit, so an app can run on brand before anyone has set the brand up by hand.

Synthesize a unified brand style preset from a set of trained exemplars.

Compose a titled, sectioned document from a source, grounded in that source and steered by a brief. What comes out is a neutral document any export format can compile.

Join an approved Content Catalyst treatment with its finished images, charts, extracted copy, and citations, then file one responsive, on-brand landing page in your Gallery. It does not publish the page.

Design a sell sheet, one-pager, case study, report, brochure, or slide deck as real pages of selectable type, then file the editable piece and its paginated PDF.

Build the validated responsive search ad copy and exact-size square and landscape image assets for each sourced keyword group.

Check discovered destination pages and compose character-safe sitelinks using only URLs that resolve.

Turn claims extracted from the landing page into character-safe callouts that retain their source evidence.

Build structured snippets from supported lists found on the landing page, with no invented fallback values.

Build a negative-keyword list only from explicit irrelevant-intent signals found during keyword research.

Validate the complete campaign bundle and file a readable document plus structured JSON data asset in the Gallery.

Write one piece of text from a brief, a caption, a summary, a subject line, a few lines of copy, on brand and with nothing filed. The light writing step, for wiring words into the next one.

Write three alternative hook lines, body copy, a call to action, and hashtags for every clip in an approved batch, checked against each clip's real transcript, then tailor the same message to LinkedIn, X, and Instagram, each held to that network's own length. Files each as a linked item in your Gallery.

Restructure an existing document into your house style without changing a single word. Built for legal, HR, and compliance text, where a rewrite is unacceptable.

Change anything already filed in your library with a plain-language instruction: tighten a paragraph, retitle a section, restyle a page, move an element on a graphic, or push one instruction across a whole deliverable at once. The change lands in place, and the version before it stays in history to revert to.

Translate anything already filed in your Gallery into another language with the design untouched: the same layout, the same slides, the same type sizes, the same imagery, and only the words changed. The translation files as a new item beside the original, so you ship both.

Turn one finished design into a reusable template: keep the artwork, lift every word, logo, and button off it, name each fillable field the way a designer would, and learn the style and the layout. A ready template mints its own Make Job.

Read a Google Slides deck or a Figma file natively through your connected app, so its layout, type, and colors are measured from the file instead of guessed from a picture of it, ready to train as a template.

Describe what your team makes over and over and get the whole template setup back: its name, what it is for, its channel and stage, the look it takes, and exactly which parts a marketer refills on each run. It proposes only, so the answer is a definition to read and change before anything is built.

Build a template for work your team has never made before, when there is no finished design to learn from. You state what it is: the medium, the name, the stage it belongs to, the look it takes, how the parts are arranged, and exactly which fields a marketer refills on each run, whether those are the fields of a graphic or a sheet, the sections of a written piece, the beats of a script, or the shots of a film. It files the editable asset and opens the template's own Make Job, the same as one trained from a real design. When you do have a finished example, train from that instead.

Fill a trained template with new words drawn from your material, at that template's own geometry. Every production run ends here, which is why what comes out looks like what you trained.

Move a whole old deck onto a new PowerPoint template you have trained. Every slide lands on the best-fitting layout of the new design, your words carry across word for word, and the pictures come with them. What comes back is a real .pptx built from the template's own file, so every text box is still editable in PowerPoint.

Open an uploaded PowerPoint exactly as it is: one artboard per slide, every text run live, every shape and picture where the file puts them, read straight from the file with no AI step and nothing reinterpreted. The upload stays the content of record, so what you download is your deck with your edits in it and identical everywhere you did not touch it. Use it when you want THIS deck back with the ability to move a box; train a template instead when you want its design reused for new work.

The same door for a .docx: every paragraph live, read from the file's own markup with no AI step. Your tracked changes, comments, content controls, headers, footers, numbering, and styles come back untouched, because nothing we do goes near them, so the file you download is your document with your edits in it.

Take anything you already have, a report, a web page, an email, a deck, a chart, and re-set it as a press-ready printed piece at the paper size and orientation you choose. The design is redone for paper rather than converted to it: a real grid, type set in points, deliberate page breaks, running headers, tables that repeat their heads across sheets, and full color on press. Say how many pages it should come to, or leave it open and let the content decide. It files a new editable piece beside the original and exports a properly paginated PDF.

The authoring engine behind composed work: writes the finished, on-brand artifact a Job asks for, whether that is a landing page, a one-pager, a rich document, a deck, or a designed artboard, and files it as editable HTML.

Rebuild a flat image as a pixel-faithful, editable page: the imagery regenerated clean, every piece of copy a live text layer at its exact position. It exports back to PNG or PDF at the original size, so the design is no longer baked into pixels.

Compose a layered, editable graphic from a brief: one artboard, a whole deck, or a fixed full page such as a sell sheet, a one-pager, or a case study. It writes the copy, places every element on the brand's design grid, and applies brand color and type by role, so what lands is editable element by element rather than a flat picture.

Turn an approved webinar transcript and its finished clips into one grounded launch post, one mobile-first carousel, and three to five verbatim quote cards. Every deliverable is filed together in one Gallery folder, and the pack is marked Done only when every required asset exists.

The same rebuild, filed as an editable spatial graphic instead of a page, so you can drag, restyle, and swap any element on a canvas.

Take several things you already have, any mix of documents, graphics, carousels, images, video, and audio, and make ONE new thing out of them from a plain-language instruction. Five graphics become a carousel you can still edit slide by slide; a document and a photograph become a designed page. Anything it could not use comes back named, with the reason.

Turn raw numbers or a passage of data into one structured chart, a bar, a pie, a table, or a single metric, ready for the data components to render. Never invents a figure.

Analyze an uploaded design image to extract structural layout zones, margins, and content slots as a reusable preset.

Dress an HTML page in one of the ten temporal master shells, so the page arrives wearing a coherent, on-brand frame instead of bare markup.

Adapt an existing design to new copy and the workspace brand. Every word of the supplied copy is kept and checked, and the layout grows to fit it.

Design a portrait carousel one slide at a time so the slides read as one continuous piece, with PDF and image exports.

Write the display copy a graphic needs and propose several distinct visual concepts for it, then wait for you to choose one.

Produce the graphic for the concept you chose, with the exact copy, and split the editable text from the artwork so it can be changed afterwards.

Draw a diagram of a set of stages from the content that describes them.

Change one stylesheet or script of an HTML page exactly as you write it. Every other script and style in the page stays untouched, and the version before is kept in history.

Generate promo imagery from a scene description, optionally building from one or more reference images, or preserving a region you mask off.

Turn the approved image prompts in a design treatment into a consistent set of role-sized Gallery images, plus a manifest that assigns each image to its section.

Render every approved data visual in a treatment plan as a branded, source-cited SVG, file each chart in the Gallery, and return a manifest that assigns every result to its section.

Draw vector artwork from a description: an illustration, a logo mark, an icon, a seamless pattern, a badge, or a diagram. It is drawn in your exact brand colors rather than approximated, stays sharp at any size from a favicon to a trade-show banner, weighs kilobytes, and every shape in it stays editable. Give it words to set and it sets exactly those; give it none and it draws no text at all.

Change an image already filed in your library with a plain-language instruction: darken the background, remove an element, adjust the mood. The edit lands on the same image in place, and the previous version stays in history to revert to.

Take a design that arrived as a flat JPEG or PNG and lift every line of copy off it as real editable text, each at its measured position, in its own font, weight, size, and color. The picture underneath comes back as the same artwork with the lettering gone, so a design nobody kept the source file for becomes editable again.

Detect and label semantic bounding regions (headline, logo, product, CTA, person, background) on an image.

Apply targeted AI inpaint edits to designated image regions while preserving protected areas byte-for-byte.

Propose transcript-grounded webinar moments, pause for approval, then file portrait and square clips plus a copy pack for each approved moment.

Read a timecoded transcript and a target length, then select the strongest contiguous segments to keep, as an ordered cut list.

Render a finished MP4 from a cut list and its source clip, with an optional branded intro and outro, normalized to one size and frame rate.

Render every clip from an approved ClipPlan, in order: each one trimmed, cropped to BOTH a vertical 9:16 and a square 1:1 frame around wherever the speaker actually is, in your brand's colors and font with your logo, with word-synced captions burned in — then writes that clip's caption pack in the same run.

Generate a short clip, five to twenty seconds, from a supplied still image. Turns a product hero or a key visual into motion for paid social.

Perform a frame-accurate cut of a video or audio file to a specified start and end timestamp.

Read a filed, approved webinar ClipPlan, check every moment's bounds and transcript references against the current recording, then trim each approved moment.

Extract a high-resolution still frame from a video file at an exact timestamp.

Synthesize a spoken-word voiceover from a script, in the voice your organization has chosen, including a cloned brand voice.

Generate a short, licensing-safe instrumental clip from a prompt, to score a video, a reel, or a social ad.

Post one filed asset to a connected LinkedIn company page, behind a go-live approval. A scheduled list drips one post per fire; an interactive run asks before anything goes out. In a workspace where publishing is off, it is refused.

Upload an image creative into your LinkedIn Ads asset library after a human go-live approval, so it can be used in an ad. Runs only where publishing and the LinkedIn Ads connection are both switched on.

Export a filed asset as a file, a PDF, an image, Word, Slides, a web page, Markdown, plain text, video, or audio, and hand back a stable URL that serves it with no login. The same formats the Export modal offers, available as a step.

Decide how an embedded piece opens on another site: fully, or up to a section, a page count or a moment in a video with an email form to unlock the rest.

Hand a composed Google Ads campaign over as a CSV for Google Ads Editor, or as a new Google Sheet. Needs no connected ad account.

Create the approved campaign in your connected Google Ads account. It always lands paused, so nothing spends until you turn it on there.

Lay out a composed Google Ads campaign as one approval board that every push or export waits on.

Mint a stable, login-free link that shows a filed asset as a viewable page: the Share modal's link, as a step. Minting it again returns the same link.

Send one plain-language question to a connected app, your analytics, your drive, your Slack, whatever the workspace holds, and get the answer back as text you can wire into the next step. Read-only: it can look things up and run reports, never post or delete.

Tell your team something from inside a run: an in-app notification with a headline and optional detail, mirrored to your workspace's Slack destination when one is connected. Wire a step's output into the body so a scheduled run reports what it made.

Stop the run there and wait for a person. It parks as Waiting on you and continues only when someone approves, or ends with their note when they decline. The content wired into it flows through unchanged.

Wire an existing Job to run automatically whenever a platform event happens: a new source filed, or another Job finishing. The same automation Automations creates by hand, set up from inside a run.