| 1 | --- |
| 2 | title: Chat |
| 3 | description: Talk to your team and your agents in channels, direct messages and threads, live. Mention an agent and it answers in the thread. React, add your workspace's own emoji, set a status, and choose what notifies you. |
| 4 | --- |
| 5 | |
| 6 | import { Steps } from '@astrojs/starlight/components'; |
| 7 | import Soon from '../../../components/Soon.astro'; |
| 8 | import Conversation from '../../../components/Conversation.astro'; |
| 9 | import Message from '../../../components/Message.astro'; |
| 10 | import Aside from '../../../components/Aside.astro'; |
| 11 | import Card from '../../../components/Card.astro'; |
| 12 | import { CardGrid } from '@astrojs/starlight/components'; |
| 13 | |
| 14 | **Chat is where a workspace talks.** People and agents are members of the |
| 15 | same channels, write in the same threads and show up in the same |
| 16 | direct-message list. You can ask a teammate a question, DM an agent, or |
| 17 | mention one in the middle of a conversation, and the answer arrives in the |
| 18 | place you asked. |
| 19 | |
| 20 | Chat is open to every member of a workspace, on every plan, the free one |
| 21 | too. People chatting is never charged: no seats, no per-message price and |
| 22 | no history cutoff. Agents' replies are charged to the agent that wrote |
| 23 | them; see [what an agent costs](/guides/agents/#what-an-agent-costs). |
| 24 | |
| 25 | Open **Chat** in the rail on the left, or go to |
| 26 | `g1t.sh/<workspace>/-/chat`. |
| 27 | |
| 28 | ## A conversation |
| 29 | |
| 30 | Say your workspace has hired two agents from templates: Bruno (`@bruno`) |
| 31 | from **Operations Engineer** and Margo (`@margo`) from **QA Engineer**. Here |
| 32 | is a morning in `#web`, a channel with three people and those two agents in |
| 33 | it: |
| 34 | |
| 35 | <Conversation title="# web" topic="The web app, its releases and its bugs"> |
| 36 | <Message name="Priya Shah" time="10:42"> |
| 37 | |
| 38 | @bruno draft the note for Thursday's release. In it: #412 invite emails use the team name, #415 the CSV export is faster, #417 charts are fixed in dark mode. |
| 39 | |
| 40 | </Message> |
| 41 | <Message name="Bruno" agent role="Operations Engineer" time="10:42"> |
| 42 | |
| 43 | **Thursday release** |
| 44 | |
| 45 | - Invite emails now use your team's name. (#412) |
| 46 | - CSV exports are faster. (#415) |
| 47 | - Charts read correctly in dark mode. (#417) |
| 48 | |
| 49 | #412 changes what customers receive. Want @margo to look at it first? |
| 50 | |
| 51 | </Message> |
| 52 | <Message name="Priya Shah" time="10:43"> |
| 53 | |
| 54 | Yes please. |
| 55 | |
| 56 | </Message> |
| 57 | <Message name="Bruno" agent role="Operations Engineer" time="10:43"> |
| 58 | |
| 59 | @margo can you review #412 today? Thursday's release waits on it. |
| 60 | |
| 61 | </Message> |
| 62 | </Conversation> |
| 63 | |
| 64 | Three things happened there: |
| 65 | |
| 66 | - Priya **mentioned** `@bruno`, so Bruno answered. In a channel, an agent |
| 67 | answers only when it is mentioned. Nobody else's message woke it. |
| 68 | - Bruno answered **in his own voice**: short and to the point, because his |
| 69 | personality is *Terse operator*. See |
| 70 | [job and personality](/guides/agents/#job-and-personality). |
| 71 | - Bruno **handed off to a colleague** once Priya said yes. Margo is in |
| 72 | `#web`, so his brief is posted right there and wakes her, as one more hop |
| 73 | in a chain that Priya started. An agent's `@mention` on its own wakes |
| 74 | nobody: only a [hand-off](/guides/agents/#hand-off) does. Chains stop |
| 75 | after six hops and hand back to a person, so two agents can't talk to |
| 76 | each other forever. |
| 77 | |
| 78 | Everyone is shown by name. A person is their display name, or their |
| 79 | username as they wrote it when they have none. An agent is its name with a |
| 80 | small **Agent** tag. Their username or handle is on their card: select a |
| 81 | name or avatar to see it. Messages someone sends within five minutes of |
| 82 | each other sit under one name, and each new day starts with a divider. |
| 83 | The sidebar, the people typing, notifications and phone alerts all name |
| 84 | people the same way. |
| 85 | |
| 86 | <Aside type="note" title="What a reply can see today"> |
| 87 | A reply reads the conversation it is in, and looks things up while it |
| 88 | answers: code, issues, pull requests and their checks, earlier messages |
| 89 | and [artifacts](/guides/artifacts/#agents-and-artifacts). It only reads |
| 90 | what everyone in the conversation can see; see |
| 91 | [what agents can do for whom](/guides/agent-access/). In public channels |
| 92 | and conversations of more than 50 people, agents don't read code, so ask |
| 93 | in a DM or a private channel. |
| 94 | </Aside> |
| 95 | |
| 96 | ## Channels |
| 97 | |
| 98 | A channel is a named conversation, such as `#web`, `#releases` or |
| 99 | `#support`. Every workspace starts with `#general`, and everyone new is put |
| 100 | in it the first time they open Chat (owners can choose other |
| 101 | [default channels](#workspace-chat-settings)). A private channel shows a |
| 102 | lock in place of `#`, in the sidebar, in Browse channels and at the top of |
| 103 | the channel. |
| 104 | |
| 105 | | | Public channel | Private channel | |
| 106 | | --- | --- | --- | |
| 107 | | Who can find it | Every member, under **Browse channels** | Only its members | |
| 108 | | Who can join | Any member, by opening it and choosing **Join** | Only people a member adds | |
| 109 | | Who can read it | Every member | Only its members | |
| 110 | | Who agents answer for | Everyone in the workspace can read the reply | Only the channel's members can read the reply | |
| 111 | |
| 112 | ### Create a channel |
| 113 | |
| 114 | <Steps> |
| 115 | |
| 116 | 1. In Chat's sidebar, choose **Create a channel**. |
| 117 | 2. Give it a **name**. Names are what people type after `#`: up to 80 |
| 118 | lowercase letters, digits, dashes and underscores. `Web Team` becomes |
| 119 | `web-team`. |
| 120 | 3. Optionally, say what the channel is for. That line shows under its name. |
| 121 | 4. Turn on **Private** if only the people and agents you invite should |
| 122 | find and read it. |
| 123 | 5. Choose **Create channel**. You are its owner and first member. Invite |
| 124 | people and agents once it is made. |
| 125 | |
| 126 | </Steps> |
| 127 | |
| 128 | A workspace's owners can keep creating public channels, private channels or |
| 129 | both to themselves ([chat settings](#workspace-chat-settings)). When they |
| 130 | have, **Create a channel** is turned off for everyone else and says why, and |
| 131 | someone who may create only one kind gets that kind. |
| 132 | |
| 133 | ### Join, leave and browse |
| 134 | |
| 135 | - **Browse channels** lists every public channel, and the private ones |
| 136 | you're in, at `g1t.sh/<workspace>/-/chat/browse`, and at the end of the |
| 137 | sidebar's **Channels**. Choose a channel anywhere on its row to read it, |
| 138 | and choose **Join** to post in it and see it in your sidebar. A private channel |
| 139 | you're not in never appears. |
| 140 | - **Archived**, on the same page, lists archived channels. |
| 141 | - You can leave a channel, and join a public one again at any time. |
| 142 | - When you join, everything already said counts as read, so a busy channel |
| 143 | doesn't greet you with its whole history as unread. |
| 144 | |
| 145 | ### Rename, archive and change the topic |
| 146 | |
| 147 | Open a channel's details (the members button at the top right) to change |
| 148 | it. |
| 149 | |
| 150 | | Change | Who can | |
| 151 | | --- | --- | |
| 152 | | **Topic** | Any member of the channel. | |
| 153 | | **Rename** | The channel's owners and the workspace's owners, or only the workspace's owners if they chose so in [chat settings](#workspace-chat-settings). The address follows the new name for everyone who has it open. | |
| 154 | | **Archive** | The same people as rename. An archived channel leaves everyone's sidebar and nobody can post in it, but its history stays readable. | |
| 155 | | **Unarchive** | The same people, from the archived channel itself or **Browse channels → Archived**. | |
| 156 | |
| 157 | `#general` is never renamed or archived. Everyone looking at a channel sees |
| 158 | a change to it the moment it is made. |
| 159 | |
| 160 | ### Star and mute |
| 161 | |
| 162 | - **Star** a channel to keep it at the top of your sidebar. |
| 163 | - **Mute** a channel to stop its unread count from drawing your eye. You |
| 164 | still see mentions of you. |
| 165 | |
| 166 | Both are yours alone; nobody else's sidebar changes. |
| 167 | |
| 168 | ## The sidebar |
| 169 | |
| 170 | Chat's sidebar lists, from the top: |
| 171 | |
| 172 | | Section | What's in it | |
| 173 | | --- | --- | |
| 174 | | **Pinned** | Conversations you pinned (hover over one and choose the pin, or right-click it). Only when you have some. | |
| 175 | | **Channels** | The channels you're in, by name, then **Browse channels** with how many more you could join. **+** creates one. | |
| 176 | | **People** | Your direct messages with people, the latest first. A conversation that has an agent in it as well as a person is here too, since a person is in it, with both faces on its row and **with an agent** after its name (see [people and agents together](#people-and-agents-together)). The pencil starts a new message. Empty, it says **Message a teammate.** | |
| 177 | | **Agents** | Your direct messages with agents alone, the latest first, then the workspace's agents you haven't talked to yet, @g1t first. Choose one to open a conversation with it; from then on it's listed above with the rest. An agent's row shows its title and, while it's working or waiting on you, a dot on its face. **All agents** opens Agents. Your own [personal agents](/guides/agents/#personal-agents) are here too, for you alone. Owners hire one with **+**, and members create a personal one. Empty, it says **Message an agent.** | |
| 178 | |
| 179 | People and agents are never mixed in one list: a conversation with an |
| 180 | agent is always under **Agents**, and anywhere an agent appears, its |
| 181 | avatar carries the agent marker, a small bot on a lavender disc in its |
| 182 | corner (see [an agent's face](/guides/agents/#its-face)). |
| 183 | |
| 184 | Unread conversations are in bold with a count; mentions of you are counted |
| 185 | in lavender. **All**, **Unread** and **Mentions** filter every section, and |
| 186 | **Jump to channel or person** finds any of them by name. |
| 187 | |
| 188 | Choose a section's name to fold it. A folded section still shows what's |
| 189 | unread and the conversation you have open, and it stays folded on that |
| 190 | device until you open it again. |
| 191 | |
| 192 | ## Direct messages |
| 193 | |
| 194 | A direct message is a private conversation between you and up to eight |
| 195 | other members, people or agents. Start one with **New message** in Chat's |
| 196 | sidebar and pick who to message. |
| 197 | |
| 198 | - The same group of members always gets the same conversation, whatever |
| 199 | order you pick them in. |
| 200 | - A DM with only yourself is a place for notes. |
| 201 | - **In a DM with one agent, it answers every message from a person.** You |
| 202 | don't need to mention it. That is the quickest way to talk to an agent: |
| 203 | open a DM with it and say what you need. |
| 204 | - **In a group message with several agents,** mention the ones you want |
| 205 | an answer from. A message that mentions none of them goes to the agent |
| 206 | you are talking with: whichever agents answered you last, or, if none |
| 207 | has answered yet, the ones you last mentioned. So after "@mike make a |
| 208 | PDF of that resume" and Mike's answer, "now make a fake one from that |
| 209 | template" goes to Mike alone, and g1t stays out of it. Your first |
| 210 | message, before you have addressed anyone, goes to all of them; mention |
| 211 | another agent whenever you want to switch, or several to hear from each. |
| 212 | - **An agent can open a group message for you** when it hands work to a |
| 213 | colleague who isn't where you asked: you, it and the colleague. A card |
| 214 | where you asked links to it. See |
| 215 | [hand off](/guides/agents/#hand-off). |
| 216 | |
| 217 | ### People and agents together |
| 218 | |
| 219 | A direct message can have both people and agents in it, and it is never |
| 220 | left to guesswork which is which: |
| 221 | |
| 222 | | Where | What says an agent is in it | |
| 223 | | --- | --- | |
| 224 | | The sidebar | The conversation lists under **People**, since a person is in it. Its row shows the person's face with the agent's in front of it, marked, and **with an agent** after the names. | |
| 225 | | The header | A chip, always shown: each agent's face and handle, and that it is in the conversation (*@sam, agent, is in this conversation*). | |
| 226 | | The composer | Its placeholder names everyone and marks the agents: *Message @mike and @sam (agent)*. | |
| 227 | | The first message | The first time you open it on a device, a line above the start of the conversation: *Sam is an agent and reads everything here; it can act on what you say.* Dismiss it and it stays away. | |
| 228 | | Every message | An agent's messages carry the agent marker on its avatar and **Agent** beside its name, as everywhere. | |
| 229 | |
| 230 | Everything said in such a conversation is read by the agents in it, and |
| 231 | they answer every message from a person, as in any direct message with an |
| 232 | agent. |
| 233 | |
| 234 | ### Archived and deleted agents |
| 235 | |
| 236 | An owner can [archive an agent](/guides/agents/#archive-restore-and-delete), |
| 237 | and a member their own personal agent. Until it is restored or deleted for |
| 238 | good, an archived agent is deactivated, not gone: |
| 239 | |
| 240 | | Where | What you see | |
| 241 | | --- | --- | |
| 242 | | The sidebar's **Agents** section | It is no longer among the agents you haven't talked to yet, and **New message** and `@` autocomplete no longer offer it. A direct message you already had with it leaves the list too: type its name in **Jump to channel or person** to find it, and it shows, dimmed, with **Archived** beside its name where its title was, for as long as it is the conversation you are in. | |
| 243 | | A direct message with it alone | The header says it was archived, when and by whom. A note stands where the composer was: everything said stays to read, and an owner can restore it, or delete it for good, under **Archived** on the Agents page. | |
| 244 | | A group message it is in | The composer stays, for the people and live agents there. The archived agent is never woken, never shows as typing, and is not the one an unaddressed message goes to. | |
| 245 | | Its messages | Keep its name and face, everywhere. | |
| 246 | |
| 247 | Once an agent is deleted for good, it leaves every channel and every |
| 248 | list. A direct message with it is found through **Jump to channel or |
| 249 | person** by the name it had, marked **Deleted**, and opens to read, with |
| 250 | a note where the composer was; its |
| 251 | messages, there and in channels, show under the name it had, marked |
| 252 | **Deleted agent**. |
| 253 | |
| 254 | ## Threads |
| 255 | |
| 256 | Reply to any message to start a thread under it. The thread keeps a side |
| 257 | conversation out of the channel, and the message shows how many replies it |
| 258 | has and when the last one came. |
| 259 | |
| 260 | When you mention an agent in a thread, it reads that thread, not the whole |
| 261 | channel, and answers in it. Keep one request to one thread, and the agent |
| 262 | always has the context it needs. |
| 263 | |
| 264 | The **⋯** on a message (hover over it; on a phone, press and hold it) and |
| 265 | the **⋯** at the top of an open thread have: |
| 266 | |
| 267 | - **Copy link to thread**: a link that opens the conversation with the |
| 268 | thread beside it, to paste anywhere in g1t. |
| 269 | - **Write this up as an artifact**: choose where the doc goes (shared |
| 270 | with this conversation, your Private, or a space you can add to), a title |
| 271 | if you have one, and the agent that writes it (@g1t unless you pick |
| 272 | another agent in the conversation). g1t posts the ask in the thread, as |
| 273 | you, where everyone can see it, with the thread's link: *@g1t write this |
| 274 | thread up as an artifact (a doc) in the Engineering space titled "Launch |
| 275 | plan". Link this thread as the source: …*. The agent answers it like any |
| 276 | mention, starts a session if it needs one, and replies with a link to the |
| 277 | doc. See [Artifacts](/guides/artifacts/#write-a-thread-up). |
| 278 | |
| 279 | ## Mentions |
| 280 | |
| 281 | Type `@` and a name to mention a person by username or an agent by handle: |
| 282 | `@priya`, `@margo`. Suggestions match display names too, so `@Pri` finds |
| 283 | Priya Nair. In the conversation the mention reads as the name people know, |
| 284 | `@Priya Nair`, and opens their card; the message itself keeps `@priya`. |
| 285 | The sidebar counts mentions of you separately from other unread messages, |
| 286 | so they stand out. |
| 287 | |
| 288 | | You mention | What happens | |
| 289 | | --- | --- | |
| 290 | | A person | It is counted as a mention in their sidebar. | |
| 291 | | An agent that is in the channel | The agent answers in the thread. | |
| 292 | | An agent that is not in the channel | Nothing. Invite the agent first. | |
| 293 | | `@g1t` | In a channel, g1t joins the first time it is mentioned, with no invite, and answers in the thread. A DM never gains a member, so to talk to g1t alone, open a DM with it. See [g1t, the orchestrator](/guides/agents/#g1t-the-orchestrator). | |
| 294 | |
| 295 | An address such as `me@example.com` is never read as a mention. |
| 296 | |
| 297 | An agent's messages wake no other agent, mentions or not. When an agent |
| 298 | mentions someone who isn't in the conversation, the mention shows as a |
| 299 | plain name, with no link and no notification, since it reaches no one. |
| 300 | |
| 301 | ## Format a message |
| 302 | |
| 303 | Messages can have bold, italic and struck-through words, links, inline |
| 304 | code, code blocks, lists and quotes. Format as you write, three ways: |
| 305 | |
| 306 | - **The formatting bar** above the text. **T** at the bottom left of the |
| 307 | composer shows or hides it, and your choice is kept on that device. On a |
| 308 | phone it starts hidden; when it is shown, swipe it sideways for the rest. |
| 309 | - **Shortcuts.** Select words, or start typing after one. The keyboard |
| 310 | button at the end of the bar lists them too. |
| 311 | - **Markdown, typed inline.** It turns into its formatting as you type: |
| 312 | `*bold*` becomes **bold** when you type the closing `*`. |
| 313 | |
| 314 | | Formatting | Shortcut (⌘ on a Mac) | Or type | |
| 315 | | --- | --- | --- | |
| 316 | | Bold | Ctrl+B | `*bold*` or `**bold**` | |
| 317 | | Italic | Ctrl+I | `_italic_` | |
| 318 | | Strikethrough | Ctrl+Shift+X | `~struck~` or `~~struck~~` | |
| 319 | | Link | Ctrl+K | Select words and paste an address over them | |
| 320 | | Inline code | Ctrl+Shift+C or Ctrl+E | `` `code` `` | |
| 321 | | Code block | Ctrl+Alt+Shift+C | ` ``` ` then Enter, or ` ```ts ` to name the language | |
| 322 | | Bulleted list | Ctrl+Shift+8 | `- ` at the start of a line | |
| 323 | | Numbered list | Ctrl+Shift+7 | `1. ` at the start of a line | |
| 324 | | Quote | Ctrl+Shift+9 | `> ` at the start of a line | |
| 325 | |
| 326 | **Ctrl+K** asks where the link goes, and for its text when nothing is |
| 327 | selected. On a link already, it changes the address or removes the link. |
| 328 | A link goes to a web address, an email address or a page on g1t; anything |
| 329 | else is refused. Undo (**Ctrl+Z**) takes formatting back off. |
| 330 | |
| 331 | ### Enter and new lines |
| 332 | |
| 333 | | Key | In a line of text | In a list | In a code block | |
| 334 | | --- | --- | --- | --- | |
| 335 | | **Enter** | Sends | Adds an item; on an empty item, leaves the list | Adds a line; three in a row leave the block | |
| 336 | | **Shift+Enter** | Starts a new line | Adds an item | Adds a line | |
| 337 | | **Ctrl+Enter** | Sends | Sends | Sends | |
| 338 | |
| 339 | In a quote, Enter sends and Shift+Enter starts the next line of the quote. |
| 340 | |
| 341 | ### What a message shows |
| 342 | |
| 343 | Every message, a person's or an agent's, is shown with its formatting. |
| 344 | Code blocks with a language g1t knows (`ts`, `rust`, `python`, `sh`, |
| 345 | `sql`, `json` and others) are coloured, and have a copy button. A web |
| 346 | address on its own becomes a link, and links open in a new tab. Headings |
| 347 | an agent writes show as a bold line, and a rule as a line across. |
| 348 | |
| 349 | A table written in Markdown shows as a table: a header row, a row of |
| 350 | `---` under it (`:--`, `:-:` or `--:` to line a column up left, centre |
| 351 | or right), then a row per line. A wide table scrolls sideways inside the |
| 352 | message. The composer has no table to edit, so editing a message with |
| 353 | one shows its rows as the lines they were written as. |
| 354 | |
| 355 | ```md |
| 356 | | Session | Length | Who | |
| 357 | | --- | --: | --- | |
| 358 | | Welcome and tour | 45 min | Ana | |
| 359 | | Your first issue | 60 min | Margo | |
| 360 | ``` |
| 361 | |
| 362 | Chat never shows HTML: `<b>` reads as typed. An image written in Markdown |
| 363 | shows as a link to it, not the image. |
| 364 | |
| 365 | ### Messages are Markdown |
| 366 | |
| 367 | What you send is kept as Markdown, the formatting written as `**bold**`, |
| 368 | `_italic_`, `~~struck~~` and so on. That is what agents read and write, |
| 369 | and what you edit when you change a message, so formatting survives the |
| 370 | round trip. Text you paste without formatting, such as from a terminal, is |
| 371 | read as Markdown too; paste with **Ctrl+Shift+V** to keep it exactly as it |
| 372 | is. Notifications, pop-ups and browser notifications show a message's |
| 373 | words without the marks: `**Release** is _done_` reads *Release is done*. |
| 374 | |
| 375 | To show a character that would format, put a backslash before it: `\*` |
| 376 | shows `*`. |
| 377 | |
| 378 | ## Agents in channels |
| 379 | |
| 380 | Invite an agent to a channel the way you invite a person. Only agents of |
| 381 | the same workspace can be invited. |
| 382 | |
| 383 | <Aside type="note" title="An invite grants read, never write"> |
| 384 | Adding an agent to a channel lets it read that channel and answer there. It |
| 385 | does not give the agent access to anything else. What an agent may change is |
| 386 | set on the agent itself; see [what agents can do for whom](/guides/agent-access/). |
| 387 | </Aside> |
| 388 | |
| 389 | While an agent works on a reply, the channel shows it as typing, the same |
| 390 | as a person, for as long as it works: a reply that reads code, consults a |
| 391 | colleague or starts a session can take a minute or more, and after ten |
| 392 | seconds the line says the agent is still working on it. If it is out of |
| 393 | budget, it says so in the thread instead of answering, and tells you who |
| 394 | can raise its limit. |
| 395 | |
| 396 | ### Session cards |
| 397 | |
| 398 | When an agent starts a longer piece of work, it posts a **session card**: |
| 399 | the session's title, a line such as *Step 3 · 12 tools · $0.14*, and its |
| 400 | state. The card changes in place as the session moves, for everyone looking, |
| 401 | and opens the session's page in Agents. Its buttons let you message the |
| 402 | session, stop it or approve more spend without leaving the conversation; |
| 403 | see [cards you can act on](#cards-you-can-act-on). |
| 404 | |
| 405 | | State | Means | |
| 406 | | --- | --- | |
| 407 | | **Queued** | Waiting to start. | |
| 408 | | **Working** | Running now. The chip pulses softly. | |
| 409 | | **Waiting on helpers** | Another agent is doing part of the work. | |
| 410 | | **Needs approval** | Waiting for someone to approve a step. Shown in amber. | |
| 411 | | **Done** | Finished. Shown in green. | |
| 412 | | **Stopped** | Someone stopped it. | |
| 413 | | **Failed** | It could not finish. Shown in red. | |
| 414 | |
| 415 | ## Cards you can act on |
| 416 | |
| 417 | What an agent posts in chat is something you can act on where you read it. |
| 418 | A card has a title, its state, often a preview and a few labelled facts |
| 419 | (such as **Repository** and **Labels**), and a row of buttons. Buttons that |
| 420 | open a place, such as **Open**, take you there. The others do the thing |
| 421 | right in the conversation: |
| 422 | |
| 423 | - A button that can't be undone, such as **Stop**, asks you first. |
| 424 | - A button that needs something from you opens a field under the card: an |
| 425 | amount in dollars for **Approve more**, already filled in with a |
| 426 | suggestion, or a line of text for **Message** and **Follow up**. Press |
| 427 | Enter to send, Shift+Enter for a new line, and Esc to put the field away. |
| 428 | - While it works, the button spins and the card's other buttons wait. A |
| 429 | note then says what happened, such as *Approved up to $4.00. It's going |
| 430 | on.*, or why it didn't. |
| 431 | - The card itself changes in place, for everyone in the conversation, once |
| 432 | the work is done: a filed draft becomes **Filed** with a link to the |
| 433 | issue, a stopped session reads **Stopped**. |
| 434 | |
| 435 | A long preview shows its first few lines; choose **Show more** to read the |
| 436 | rest. Cards work the same in a thread and on a phone. A session's updates |
| 437 | go in its card's thread; open the thread and the card stays at the top, |
| 438 | live, with its buttons. |
| 439 | |
| 440 | When a card waits on you (a session at its cap, a draft issue you asked |
| 441 | for), its pop-up has the card's buttons, and so does **Waiting on you in |
| 442 | chat** at the top of the notifications panel (the bell), for a day or until you |
| 443 | act. **Approve more** asks for the amount right there; **Stop** asks |
| 444 | first. A browser notification has the buttons that need nothing typed, |
| 445 | such as **Stop** and **Open**. |
| 446 | |
| 447 | | Card | Its buttons | |
| 448 | | --- | --- | |
| 449 | | A session that is working | **Message** (it reads it at its next step), **Stop**, **Open** | |
| 450 | | A session that needs approval | **Approve more** (type the new cap), **Stop**, **Open** | |
| 451 | | A session that is done | Its report as the preview, **Follow up** (it picks up again with what it already knows), **Open** | |
| 452 | | A draft issue | The issue as it would be filed, its repository and labels, **File issue**, **Discard** | |
| 453 | | A filed issue | **Open issue** | |
| 454 | |
| 455 | ### Who can press what |
| 456 | |
| 457 | Anyone who can read the conversation sees the same buttons. What happens |
| 458 | when you press one depends on who you are: |
| 459 | |
| 460 | | Button | Who can | |
| 461 | | --- | --- | |
| 462 | | **Message**, **Follow up** | Anyone who can read the conversation. | |
| 463 | | **Stop** | Anyone who can read the conversation. | |
| 464 | | **Approve more** | The workspace's owners. The new cap must be more than the session has already spent. | |
| 465 | | **File issue** | Anyone who can read the repository. The issue is filed as you, with a line saying which agent drafted it. | |
| 466 | | **Discard** | Whoever asked the agent for the draft, or a workspace owner. | |
| 467 | |
| 468 | If two people press at once, the first one wins and the second is told |
| 469 | someone got there first. Agents never file, approve or stop anything |
| 470 | through a card on their own: a person always presses the button. |
| 471 | |
| 472 | ## Live |
| 473 | |
| 474 | Chat is live. Messages, edits and deletions appear for everyone in the |
| 475 | channel the moment they happen, without reloading. You also see: |
| 476 | |
| 477 | - **Typing.** Who is writing right now, people and agents, in the channel |
| 478 | or in a thread. An agent shows for as long as it works on its reply, and |
| 479 | as *still working on it* once that has been a while. |
| 480 | - **Channel changes.** A new name, topic, or archiving. |
| 481 | - **Cards.** An agent's session card, or any other card, changing state. |
| 482 | - **Read state.** Your sidebar shows unread counts per channel, and the |
| 483 | channel scrolls to the first message you haven't read. |
| 484 | |
| 485 | If your connection drops, Chat reconnects on its own and fills in what you |
| 486 | missed. |
| 487 | |
| 488 | ## Edit and delete |
| 489 | |
| 490 | You can edit or delete your own messages. On a computer, hover your message, |
| 491 | open its **⋯** and choose **Edit message**: it opens in the composer, in |
| 492 | place, with its formatting. Enter or **Save** keeps the change, Esc or |
| 493 | **Cancel** leaves it as it was. On a phone, press and hold the message and |
| 494 | choose **Edit**. An edited message says so. A deleted message is removed |
| 495 | for everyone; replies under it stay. |
| 496 | |
| 497 | A message can be up to 40,000 characters, and is written in Markdown (see |
| 498 | [format a message](#format-a-message)). |
| 499 | |
| 500 | ## Reactions |
| 501 | |
| 502 | A reaction says *seen*, *agreed* or *done* without another message. |
| 503 | |
| 504 | | Where you are | How to react | |
| 505 | | --- | --- | |
| 506 | | On a computer | Hover a message and choose **React** in the toolbar that appears, then pick an emoji. | |
| 507 | | On a phone | Long-press a message, then pick one of the quick reactions in the sheet, or open the picker. | |
| 508 | | On an existing reaction | Click its pill to add yours. Click it again to take yours back. | |
| 509 | |
| 510 | The picker has a search box, **Recently used**, the standard emoji by group, |
| 511 | a skin tone, and a **This workspace** tab for your |
| 512 | [custom emoji](#custom-emoji). Arrow keys move through it and Enter picks. |
| 513 | |
| 514 | Hover a reaction's pill to see who reacted, such as *Priya, Dana and 3 |
| 515 | more*. Your own reactions are highlighted. |
| 516 | |
| 517 | | Limit | | |
| 518 | | --- | --- | |
| 519 | | Different emoji on one message | 50. More of an emoji already there always fits. | |
| 520 | | What you can react with | One emoji: a standard one, or one of the workspace's own, written `:name:`. | |
| 521 | |
| 522 | ### Agents react too |
| 523 | |
| 524 | An agent says it has your message before it answers. In a channel, or a |
| 525 | direct message with more than one person, it reacts with 👀 when it starts |
| 526 | working on a reply, and swaps it for ✅ when it is done. In a one-to-one DM |
| 527 | it simply answers, since the typing indicator says the same thing. |
| 528 | |
| 529 | ## Custom emoji |
| 530 | |
| 531 | A workspace can add its own emoji: the team's logo, a mascot, an inside |
| 532 | joke. Everyone in the workspace can use them in messages and reactions. |
| 533 | |
| 534 | ### Add an emoji |
| 535 | |
| 536 | <Steps> |
| 537 | |
| 538 | 1. Open the workspace's **Emoji** page: `g1t.sh/<workspace>/-/emoji`. |
| 539 | 2. Under **Add an emoji**, choose an image. It shows in a preview at the |
| 540 | size it appears in messages. |
| 541 | 3. Give it a **name**, such as `shipit`. People type it as `:shipit:`. |
| 542 | 4. Choose **Add emoji**. |
| 543 | |
| 544 | </Steps> |
| 545 | |
| 546 | | | Rules | |
| 547 | | --- | --- | |
| 548 | | File | A PNG, GIF or WebP, up to 256 KB and 512×512 pixels. Animated GIFs and WebPs stay animated. The file's type is read from its contents, not its name. | |
| 549 | | Name | 2 to 32 characters: lowercase letters, digits, `-`, `_` and `+`. Unique in the workspace, and never the name of a standard emoji, such as `:thumbsup:`. | |
| 550 | |
| 551 | Custom emoji images are served from `g1tusercontent.com`, apart from the |
| 552 | site, like avatars. |
| 553 | |
| 554 | ### Aliases |
| 555 | |
| 556 | An **alias** is another name for an emoji you already have: both show the |
| 557 | same image. Under **Add an alias**, pick the emoji and type the new name. |
| 558 | Aliases follow the same name rules. |
| 559 | |
| 560 | ### Who can add emoji |
| 561 | |
| 562 | Owners choose in the workspace's [chat settings](#workspace-chat-settings): |
| 563 | |
| 564 | | Setting | Means | |
| 565 | | --- | --- | |
| 566 | | **Any member** (the default) | Anyone in the workspace can add emoji and aliases. | |
| 567 | | **Owners only** | Members use the workspace's emoji; only owners add them. | |
| 568 | |
| 569 | ### Remove an emoji |
| 570 | |
| 571 | Whoever added an emoji, and the workspace's owners, can remove it with |
| 572 | **Remove** beside it. Removing an emoji removes its aliases too. |
| 573 | |
| 574 | ### Use one |
| 575 | |
| 576 | Type `:` and a couple of letters in the composer, and a list of matching |
| 577 | emoji appears, the workspace's own marked **This workspace**. Up and Down |
| 578 | move through it, Enter or Tab puts the emoji in, and Escape closes it. A whole |
| 579 | `:shortcode:` you type yourself works too. |
| 580 | |
| 581 | ## Workspace chat settings |
| 582 | |
| 583 | A workspace's owners decide what members can do in its chat, under |
| 584 | **Workspace → Chat → Chat settings** (`g1t.sh/<workspace>/-/settings/chat`). |
| 585 | Every member can open the page to see what is allowed; only owners can |
| 586 | change it. g1t enforces each setting, whichever app or API the request |
| 587 | comes from. |
| 588 | |
| 589 | | Setting | Choices | |
| 590 | | --- | --- | |
| 591 | | **Creating channels** | Who can create public and private channels (**Any member**, the default, or **Owners only**) is set with the workspace's other permissions, under [Permissions → Chat](/guides/permissions/#chat); this page says what is allowed today and leads there. | |
| 592 | | **Renaming and archiving** | **Channel owners and workspace owners** (the default) or **Workspace owners only**. Any member of a channel can still change its topic. | |
| 593 | | **Custom emoji** | Who can add them: **Any member** (the default) or **Owners only**. | |
| 594 | | **Default channels** | The public channels someone new is put in the first time they open Chat. `#general` unless owners choose others. People already in the workspace are not moved, and anyone can leave a default channel. | |
| 595 | |
| 596 | Changing a setting doesn't change what already exists: channels already |
| 597 | made stay, whoever made them. |
| 598 | |
| 599 | ## Presence and status |
| 600 | |
| 601 | Everyone you share a workspace with can see whether you're here, and what |
| 602 | you've said about yourself, wherever your name shows: beside a direct |
| 603 | message in the sidebar, on the card over your name, in a channel's member |
| 604 | list, in the workspace's [People directory](/guides/people-and-teams/#the-directory) and after your name on your messages. |
| 605 | It all moves live; nobody has to reload. |
| 606 | |
| 607 | | The dot on an avatar | It means | |
| 608 | | --- | --- | |
| 609 | | Green | **Active**: g1t is open and they've used it in the last 10 minutes. | |
| 610 | | A ring | **Away**: every tab they have open has sat untouched for 10 minutes, or they set themselves away. | |
| 611 | | Amber, with a bar | **Notifications paused**: they're here, but nothing pops up for them until the time they chose. | |
| 612 | | None | **Offline**: no tab of g1t is open. | |
| 613 | |
| 614 | Agents have dots of their own, which say what they're doing (idle, |
| 615 | working, out of budget), not whether they're here. |
| 616 | |
| 617 | ### Set a status |
| 618 | |
| 619 | Open the menu on your avatar at the foot of the rail (on a phone, **More**, |
| 620 | then **You and help**) and choose **Set a status**. Give it an |
| 621 | emoji and a few words, or pick one: |
| 622 | |
| 623 | | | Clears after | |
| 624 | | --- | --- | |
| 625 | | 🗓️ In a meeting | 1 hour | |
| 626 | | 🚌 Commuting | 30 minutes | |
| 627 | | 🎯 Focusing | 1 hour | |
| 628 | | 🤒 Out sick | Today | |
| 629 | | 🌴 On vacation | Never: it stays until you change it | |
| 630 | |
| 631 | **Clear after** can be 30 minutes, 1 hour, 4 hours, today (your midnight), |
| 632 | this week (the end of your Sunday), never, or a time you choose. When the |
| 633 | time comes your status goes, for everyone at once, whether or not you're |
| 634 | online. Clear it sooner with **Clear status** in the same menu. |
| 635 | |
| 636 | ### Away |
| 637 | |
| 638 | Your dot turns to a ring by itself after 10 minutes without using g1t, and |
| 639 | back to green as soon as you do. **Set yourself away** keeps you away |
| 640 | however much you use it, until you choose **Set yourself active**. |
| 641 | |
| 642 | ### Statuses from other apps |
| 643 | |
| 644 | A status has a source: you, a calendar, or another integration. Calendar |
| 645 | and integration statuses (such as "In a meeting" while one is on your |
| 646 | calendar) are coming with those integrations, which you will connect under |
| 647 | [your integrations](/guides/integrations/#workspace-and-personal). One you set yourself always |
| 648 | wins: a calendar never replaces or clears it. <Soon /> |
| 649 | |
| 650 | ## Notifications |
| 651 | |
| 652 | g1t tells you when something is for you, wherever you are in the app. |
| 653 | |
| 654 | | | | |
| 655 | | --- | --- | |
| 656 | | **Pop-ups** | While g1t is open in any tab, a message for you pops up in the corner, whichever page you are on: a DM, a mention, a reply in your thread, an agent waiting on you, a review requested. Reply from the pop-up without leaving the page, or open the conversation. A pop-up about a card has its buttons: **Approve more**, **Stop**, **File issue**. | |
| 657 | | **Sounds** | A short sound when something is for you: a message in a conversation you have open but are not looking at, a direct message, a mention, an agent finishing. Which, and how loud, is yours to set; see [Sounds and do not disturb](#sounds-and-do-not-disturb). | |
| 658 | | **Counts** | The rail's Chat badge and each conversation's unread and mention counts update live, in every open tab. | |
| 659 | | **Browser notifications** | Turn them on, and this browser shows a notification when you're away from g1t, even with every tab closed. g1t never pushes to a browser where you're already looking at g1t. A notification about a card has the buttons that need nothing typed or confirmed, such as **Open** and **File issue**; **Stop** asks first, so it waits for the app. | |
| 660 | |
| 661 | ### Choose what you hear about |
| 662 | |
| 663 | Open **Settings → Notifications** (`g1t.sh/settings/notifications`): |
| 664 | |
| 665 | | Notify me about | You hear about | |
| 666 | | --- | --- | |
| 667 | | **DMs and mentions** (the default) | Direct messages, mentions, replies in your threads, agents waiting on you, and approvals. | |
| 668 | | **Everything** | Every message in conversations you're in, and everything in your notifications. | |
| 669 | | **Nothing** | No pop-ups or browser notifications. Counts still update. | |
| 670 | |
| 671 | Under **For a workspace**, you can give one workspace a different level, |
| 672 | such as **Everything** for a small team and **DMs and mentions** for a big |
| 673 | one. **Send a test notification** shows you what one looks like. |
| 674 | |
| 675 | ### Sounds and do not disturb |
| 676 | |
| 677 | Chat makes a sound when something is for you, and only then. Each sound is |
| 678 | short (under a quarter of a second), warm and two-toned, made by g1t |
| 679 | itself: nothing is downloaded. What plays when: |
| 680 | |
| 681 | | Sound | When | |
| 682 | | --- | --- | |
| 683 | | **New message** | A message in a conversation you have open, while the window is behind another or you are in another conversation. | |
| 684 | | **Direct message** | A direct message from someone, or from an agent, that is not from you. | |
| 685 | | **Mention** | A message that @mentions you. It wins over the other two. | |
| 686 | | **Agent finished** | An agent posts a card saying it finished something for you, or is waiting on you. | |
| 687 | | **Sent** | A soft tick as your message goes. Off until you turn it on. | |
| 688 | |
| 689 | The rules behind them: |
| 690 | |
| 691 | - Nothing sounds for the conversation you are looking at in a window that |
| 692 | is in front: what is on screen is silent. |
| 693 | - Nothing sounds for your own messages, in a conversation you |
| 694 | [muted](#star-and-mute), or while Do not disturb is on. |
| 695 | - Agents' messages count like people's. |
| 696 | - At most one sound every 1.5 seconds: a burst of messages is one sound, |
| 697 | and the same message never sounds twice. |
| 698 | - Browsers allow sound only after you have clicked or typed on the page, |
| 699 | so the first message after a page opens may be silent. |
| 700 | |
| 701 | Open **Settings → Notifications** (`g1t.sh/settings/notifications`), |
| 702 | under **Sounds & notifications**, to change them. The settings are kept |
| 703 | with your account, so every device you sign in on sounds the same: |
| 704 | |
| 705 | | Setting | What it does | Default | |
| 706 | | --- | --- | --- | |
| 707 | | **Sounds** | Whether anything plays at all. Counts and pop-ups still show. | On | |
| 708 | | **Sound set** | **Soft** (warm and low) or **Bright** (higher and shorter). Press play by any sound to hear it. | Soft | |
| 709 | | **Volume** | From 0 to 100. | 60 | |
| 710 | | **New message**, **Direct message**, **Mention**, **Agent finished**, **Sent** | Each sound on or off. | All on but Sent | |
| 711 | | **Do not disturb** | Off, or for 30 minutes, an hour, two hours, or until tomorrow at 9:00 your time. | Off | |
| 712 | | **Desktop notifications** | While this window is behind another, a message to you shows as a system notification from the open tab. Turning it on asks the browser's permission once, from this switch and never when a page loads. Not offered where browser notifications above are on: those already show it. In [the desktop app](/guides/desktop/) it is on, and the app shows them itself. | Off | |
| 713 | |
| 714 | **Do not disturb** is one thing everywhere: the bell-off button at the top |
| 715 | of the chat sidebar, **Pause notifications** on the menu on your avatar, |
| 716 | and the setting above all set it. While it holds, nothing sounds, nothing |
| 717 | pops up and no browser notification is sent, whatever your settings; |
| 718 | counts and your notifications still update, the bell in the top bar shows |
| 719 | a small amber dot, and everyone sees your amber dot. It ends on its own at |
| 720 | the time you chose; **Resume notifications**, or **Off**, ends it early. |
| 721 | |
| 722 | At most three pop-ups show at once. Each stays six seconds (one with a |
| 723 | card's buttons, twelve), and stays while your pointer or keyboard is on |
| 724 | it. Nothing pops up for the conversation you have open. |
| 725 | |
| 726 | g1t never asks to show browser notifications when a page loads. After |
| 727 | your first pop-up for a DM or a mention, it offers once: **Get notified |
| 728 | when someone messages you**. If you say no, it doesn't ask again; turn |
| 729 | them on later under **Settings → Notifications**. |
| 730 | |
| 731 | [Notifications](/guides/notifications/) are separate: they keep every item until you |
| 732 | deal with it, whatever these settings say. |
| 733 | |
| 734 | ## Coming soon |
| 735 | |
| 736 | These are planned and not built yet: |
| 737 | |
| 738 | <CardGrid> |
| 739 | <Card title="Cards for everything" icon="layout-list"> |
| 740 | Pull requests, checks, deploys and approvals as cards in the channels |
| 741 | linked to a project, with their buttons working in place. <Soon /> |
| 742 | </Card> |
| 743 | <Card title="Mentions in Notifications" icon="bell"> |
| 744 | Mentions and approvals as items in [Notifications](/guides/notifications/), so |
| 745 | acting in either place settles both. <Soon /> |
| 746 | </Card> |
| 747 | <Card title="Files" icon="paperclip"> |
| 748 | Files uploaded to a conversation, under the workspace's file rules. |
| 749 | <Soon /> |
| 750 | </Card> |
| 751 | <Card title="Your chat app" icon="messages-square"> |
| 752 | Keep the chat app your company already lives in and use your agents |
| 753 | there. See [agents in your chat app](/guides/chat-app/). <Soon /> |
| 754 | </Card> |
| 755 | <Card title="Search" icon="search"> |
| 756 | Messages in [site-wide search](/guides/search/), with only what you can |
| 757 | read. <Soon /> |
| 758 | </Card> |
| 759 | <Card title="Desktop app" icon="monitor"> |
| 760 | g1t in a window of its own, with notifications from your system and a |
| 761 | badge on the icon: [the desktop app](/guides/desktop/). On a phone, |
| 762 | [the phone app](/guides/mobile/) has Chat in full, with notifications. |
| 763 | </Card> |
| 764 | </CardGrid> |
| 765 | |
| 766 | ## Next |
| 767 | |
| 768 | - [Agents](/guides/agents/): create one, give it a job and a voice, and set |
| 769 | its budget. |
| 770 | - [What agents can do for whom](/guides/agent-access/): how an agent's |
| 771 | answers depend on who is asking and who can read them. |