> For the complete documentation index, see [llms.txt](https://help.craftsmanplus.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.craftsmanplus.com/intelligence/chatbot/user-input/attachments.md). # Attachments Add images, PDFs, audio, video, fonts, and Asset Library assets to a message, and reuse media that is already in the conversation. A message does not have to be text alone. You can attach files and Asset Library assets so the assistant has something to look at, work from, or save, and you can point it back at media that already appears earlier in the conversation. *** ### Ways to Attach There are four ways to add a file to your message. They all accept the same file types and apply the same limits. **Upload files.** Click the **+** button next to the message box and choose **Upload files**. Select one or more files from your computer. **Asset Library.** Click the **+** button and choose **Asset Library**. The Asset Library opens so you can select one or more assets, then click **Attach**. **Drag and drop.** Drag files from your desktop anywhere onto the chat. The chat shows a dashed outline with **Drop to Upload** while you hover. Release to attach. **Paste.** Copy an image or file and press **⌘ V** or **Ctrl + V** in the message box. Pasting text still pastes text as usual. *** ### Supported Files | Type | Files | What the assistant can do with it | | ---------- | -------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------- | | **Images** | PNG, JPG, WebP, GIF, SVG, and other common image formats | See the image. Use it as a reference, analyze it, add it to your project, or save it to your Asset Library. | | **PDF** | PDF | Read the document. Useful for briefs and brand guidelines. | | **Audio** | MP3, WAV, AAC, OGG, FLAC, M4A, WMA, AIFF | Add it to your project or save it to your Asset Library. The assistant cannot listen to audio, so it works from the file's name and length. | | **Video** | MP4, WebM, MOV | Capture still frames from it to look at, add it to your project, or save it to your Asset Library. The assistant cannot watch video directly. | | **Fonts** | TTF, OTF, WOFF, WOFF2 | Save the font to your Asset Library so it becomes available in the editor. | Files of any other type are skipped. If none of the files you selected are supported, you will see the message **No supported files selected**. {% hint style="info" %} *Large images are automatically resized before they are sent. This keeps uploads fast and has no effect on the original file on your computer.* {% endhint %} *** ### Limits * Up to **20 files** per message. * Images, PDFs, audio, and fonts up to **50 MB** each. PDFs must also be under **20 MB** after processing. * Videos up to **100 MB** each. If a file is over the limit you will see a message naming the file, for example *"brand-deck.pdf is too large (max 20MB)"*. If you select more than 20 files at once, none of them are added, so trim the selection and try again. *** ### Before You Send Attached files appear as tiles above the message box. Images and videos show a thumbnail. PDFs, audio, and fonts show an icon and the file name. * A spinner on a tile means the file is still uploading. **Send** stays disabled until every upload has finished. * Hover over a tile and click the **X** to remove it before sending. * You can mix file types in one message, and you can send a message with attachments and no text. *** ### Reusing Media from the Conversation Any image or audio clip that already appears in the chat, whether you uploaded it or the assistant generated it, can be used again without re-uploading. Hover over an image or audio clip in a message and click the **select** toggle that appears in its corner. The item is added above the message box as a reference tile. Select as many as you need, then type your request and send. Click the toggle again, or the **X** on the tile, to deselect. As soon as you have media selected or attached, two suggested actions appear above the message box: * For images: **Add these images to my scene** and **Save these images to my asset library** * For audio: **Add this audio to my scene** and **Save this audio to my asset library** Click one to send it as your message. This is the quickest way to place a generated asset in your project or keep it for later. You can also type your own request instead, for example *"Use the second image as the background of scene 2."* {% hint style="info" %} *Clicking an image in the chat selects it as a reference. There is no full-size preview in the chat itself.* {% endhint %} *** ### Attachments in the Conversation Once sent, attachments appear with your message: * **Images** display in a grid. * **PDFs and fonts** display as a file card. Click the card to open the file in a new tab. * **Audio** displays as an inline player with play and pause. * **Video** displays as an inline player. * **Asset Library assets** display as thumbnails. Click one to open the Asset Library at that asset. Images the assistant fetches for itself, such as a screenshot of your scene or a picture from a link you shared, are not shown in the chat. *** ### Saving Attachments Uploading a file to the chat does not add it to your Asset Library. The same is true of assets the assistant generates. Nothing is saved to your library or placed in your project unless you ask. To keep something, click **Save ... to my asset library** or **Add ... to my scene** above the message box, or ask in your own words. The assistant confirms what it saved and where. See [Asset Generation](/intelligence/specialties/asset-generation.md) for how generated assets work. {% hint style="warning" %} *Generated audio and video are stored temporarily. Save any clip you want to keep to your Asset Library before you move on.* {% endhint %} *** ### Sharing Links Paste a URL into your message like any other text. The assistant can fetch the page to read its content, or, for a link to an image or PDF, fetch the file itself to look at. This is useful for pointing the assistant at a website, an app store listing, or a hosted brief. > *"Match the style of this landing page: "* *** ### Working with Your Selection You do not need to attach layers. When you have a PLAY project open, the assistant can read what you currently have selected in the layers panel. Select the layers you want to work on and refer to them directly. > *"Make the selected layers 20% larger."* > > *"Change this layer's font to match the headline."* --- # 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 by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://help.craftsmanplus.com/intelligence/chatbot/user-input/attachments.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. 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.