Modules

What can Welcome do?

With Welcome you can automatically greet new members, say goodbye to departing members and assign roles when joining. All three areas can be used independently of each other.

Required Discord rights

Fjord needs this permission only when the normal text of a welcome or farewell message should ping a role that not everyone can mention. @everyone and @here are blocked in the Welcome system and do not work there.

“Attach files” is required for the welcome card. Manage Roles is only required for automatic roles.

Setup

1. Select the server you want to manage.
2. Open Welcome.
3. Enable welcome messages, farewell messages or automatic roles.
4. Select an accessible text channel or at least one role.
5. Check the preview and save your changes.

For automatic roles, the Fjord role must be above all roles to be assigned.

Welcome message

The switch and channel selection can be edited separately. Switching off preserves the selected channel. No channel initially clears only the selection, the editor stays accessible. Saving without a channel disables that service and closes its message editor. This applies to both Welcome and Farewell.

The welcome message is sent once a member joins the server. Channel, message text and welcome card are saved together. A preview shows how text and card interact before saving.

Placeholder for messages

Use the placeholder selection in the editor. This way the name is used exactly and typos in curly brackets are avoided.

Message editors and card attachments

Without saved custom settings, welcome has the card enabled and embeds disabled. Farewell starts with normal text, without an embed or card. Welcome, farewell and join roles are activated independently. Both message editors offer V1 and V2, optional text and optional embeds. The generated-card switch is in the welcome editor. Empty message text uses the displayed default text.

Message content up to 1,800 characters, separate from embed text.

Welcome and farewell each have their own V1/V2 selector. Switching requires confirmation and resets that editor’s message text, embeds, images and link buttons. The channel, activation and separate card design are retained. Save applies the switch. Cancel restores the previously saved state.

V1

V1 uses classic Discord embeds with an accent colour and fixed sections. Up to ten embeds are supported. The table gives maximum character counts. Author, title, description, footer and every field name and value together may contain at most 6000 characters across all embeds. Normal message text has a separate limit. Each V1 embed supports up to 25 fields.

All design sections are optional. Once you add a field, both its name and value are required. Remove unused fields. Side-by-side enables inline fields, whose final layout depends on Discord and screen width. An author link or icon requires an author name, a title link requires a title and a footer icon requires footer text. Optional links must be complete HTTP(S) addresses. The timestamp shows the sending time.

Role and channel pickers are available in normal message text, the V1 description and field values. These mention pickers are unavailable in author, title, footer and field names. Server emojis can be used in those text sections. Author URL and title link are URL-only inputs without roles, channels or emojis. Mentions in classic embeds do not notify anyone.

Discord only shows the author icon when an author name is also set. Enter a name above so the icon appears in the preview. Discord only shows the footer icon when footer text is also set. Enter text above so the icon appears in the preview.

V2

An image block can contain a gallery of up to ten images. Use “+ Another image” to add images, which can be removed individually. Each image and thumbnail has an optional description of up to 80 characters and an optional spoiler setting. A section supports up to three texts of up to 4000 characters each. Use “+ Another text” to add up to two additional texts. Accessory lets you choose between a thumbnail and a link button. The link button requires a label of up to 80 characters and a complete HTTP(S) URL of up to 512 characters. Emoji and disabling are optional. Additional section texts and the accessory button count towards the component limit. Individual images in a gallery do not count as additional components.

In V2, containers and images, including thumbnails, can be marked as spoilers. Their content is initially hidden and revealed by clicking.

V2 offers up to ten containers with separate accent colours. Use text blocks or sections with an optional thumbnail on the right, each supporting up to 4000 characters and Markdown. Separators can show a line or just spacing, with small or large spacing. Add, copy, remove and reorder blocks with the arrows. Empty text and image blocks produce no visible content. V2 text sections offer role, channel and server-emoji pickers.

V2 has no classic author fields, title-link inputs, inline fields, footer icons or timestamp switch. Add headings and links using Markdown in text blocks. Use the Markdown example below for a small footer and the time token for a timestamp. V1 does not offer freely arranged V2 blocks. Both formats cannot be mixed in one message.

-# ... · {time.full}

V2 has a shared limit of 40 components for the entire message. Containers, text displays, sections, thumbnails, image galleries, separators, action rows, buttons and select menus each count. Menu options do not count individually. Automatically added separators and controls also use the budget. For example, a container with text, an image, a separator and a button with its action row uses six components. Ten containers therefore cannot all be filled without limit.

The six message placeholders also work in the text sections of both embed formats. The username is used when no global display name exists. The member-mention placeholder is not offered in V1 sections that do not support mentions. For V1 author icons, thumbnails, large images and footer icons, and for V2 thumbnails and image blocks, choose an uploaded image, the server icon or the member avatar. A missing server icon leaves the image empty. The preview uses your account, while sent messages use the actual member. These dynamic image sources are exclusive to welcome and farewell.

With embeds disabled, the enabled welcome card is sent as a separate image with normal text. With embeds enabled, it occupies the large-image slot of the first V1 embed or is inserted in the first V2 container. That card slot cannot be overwritten with your own large image. Other embeds and additional V2 image blocks can use uploaded images. Reordering moves the card slot to the new first embed. Card design is independent of V1/V2.

Image limits apply per file. In V1, author icons, thumbnails and footer icons may each be up to 5 MB. The large-image limit matches the V2 image-block limit shown below. PNG, JPG/JPEG, WebP and GIF are supported. Choose a file to replace an image, or Remove to clear its slot. Author and footer icons are invisible without their corresponding text.

Insert content in messages

The message editor has its own selections for placeholders, roles, channels and server emojis. A selected entry is inserted at the current writing position.

You can open the selections using the buttons in the editor. While writing, @ opens the role selection, # opens the channel selection, and : opens the server emoji selection. Further characters entered filter the respective list.

Welcome and farewell messages can contain a maximum of 1,800 characters each. Inserted entries count as part of the message content.

Design welcome card

The card editor offers Content, Layout, Surface, Ring and Text.

There are no independent controls for avatar size, individual font sizes or arbitrary text positions.

Reset restores the default design. Save or Cancel uses the shared save bar.

Use the default background, a solid color or your own image. Move and enlarge an uploaded image in the crop editor. Canvas also provides a liquid layer with color, opacity and blur.

Title ≤ 120 · Subtitle ≤ 160

Card placeholders

{member.user.name} · {guild.name} · {guild.member_count}

The card is 1200 × 600 pixels and is output as a static PNG. The preview uses your account and sample details from the selected server. On join, the member's details are used. Browser previews require a valid dashboard connection.

Farewell message

The switch and channel selection can be edited separately. Switching off preserves the selected channel. No channel initially clears only the selection, the editor stays accessible. Saving without a channel disables that service and closes its message editor. This applies to both Welcome and Farewell.

The farewell message is sent when a member leaves the server. It can use a different target channel and its own text. Mentions cannot always be resolved after leaving. Use {member.user.global_name}.

Automatic roles

Fjord can assign multiple roles to new members immediately after joining. Roles managed by Discord and roles above the Fjord role cannot be assigned.

1. Check under Server Settings → Roles that the Fjord role is above the target roles.
2. Activate automatic role assignment and select the desired roles.
3. Save the selection.

Roles with extensive administrative rights should not be assigned automatically. Fjord only displays roles that are generally assignable.

Fix errors

Link buttons

V2 groups up to five link buttons into one action row. V2 uses the overall component budget instead of the V1 limit of five rows.

Link buttons open websites and do not assign roles. Each added button requires a label of up to 80 characters and a complete HTTP(S) URL of up to 512 characters. An emoji is optional. Buttons can be reordered and removed. V1 displays them below the embeds with up to five buttons per action row and five rows overall, so at most 25 buttons. V2 displays them inside their container and counts them and their action rows towards the 40-component budget.

Link buttons are sent only when embeds are enabled.

Requirements and limitations

A valid design does not require every optional section. Colour alone or empty blocks do not supply visible content for publishing. The preview does not send a message or replace testing in the target channel. Neither editor offers polls, stickers, forms or freely programmable actions. Available data and editing also depend on your granted dashboard permissions.

Preview, spoilers and image files

In the V2 preview, a single gallery image keeps its aspect ratio. Multiple images form a grid with equally sized 16:9 crops. Two or four images use two columns, while three and five to ten images use three columns. A thumbnail appears to the right of the section text. The image description provides alternative text rather than an extra visible caption.

Image and container spoiler switches belong to the V2 editor. Individual gallery images and thumbnails can be hidden independently. A container spoiler hides its contents while keeping the outer frame visible. Clicking reveals the preview without changing the saved spoiler setting. Members can also reveal spoilers by clicking them in Discord.

Message images accept PNG, JPG/JPEG, WebP and GIF. The file type must match its image contents. Renaming an HTML or SVG file does not make it a supported image. The limits in the image tables still apply to each file. If a local image is rejected or can no longer be read, select the original file again and save once more.

Markdown text areas support web and email links. The preview opens HTTP(S) and mailto links, but does not open local file or Blob links. Author links, title links and link buttons require complete HTTP(S) URLs. The preview does not publish a message. Check the saved result in Discord, where the appearance can vary by client and screen width.