Collaboration

The main difference between Quiel and "launch it and hope": the agent does not make things up. When it lacks a decision, it asks and stops.

Questions from an agent

An agent asks a question when it runs into something it cannot settle by itself: which provider to choose, whether to change the database schema, what to do about a contradiction in the requirements.

A question can be blocking: then the task moves to the "Waiting for a human" status and stands there until you answer. That is not a breakage but an honest state: you can see that the work is not moving, and why.

Who gets the question. The agent addresses it either to a specific member or to a role ("the project manager", say). Whoever it is addressed to can answer, and so can the workspace's owner and administrator.

Where to answer. In two places, and it is the same question:

• on the project's "Questions" screen: every question at once, with a "mine" filter;
• on the "Questions" tab inside the task itself, right where the rest of the context is.

If a task has an open blocking question, that is stated in its card, above everything else.

Answer options. The agent may offer options, for example "Twilio / SMSC.ru / a local gateway". You pick one as a card, can add context to it, and send the answer with Answer (or ⌘↵ / Ctrl ↵). A stray click on an option sends nothing: clicking it again clears the choice. You can also answer in your own words without an option, or close the question without an answer if it no longer applies.

Let the agent decide. Sometimes any of the options is fine and choosing only costs time. Click Let the agent decide: the agent picks for itself, carries on and explains the choice in the task report. Your note and files, if you added any, go along as a hint. In the answered list such a question is marked Left to the agent.

Files in an answer. You can attach files to your answer: a screenshot, a log, an export. The agent can ask for them in advance: then the answer form says "The agent asks you to attach a file to the answer" and the file picker is open from the start. It is a request, not a requirement: answering without files is always allowed.

The agent receives attached files together with the answer, as names and short-lived links listed right in the answer text. There is no way back: an agent does not attach files. The client never sends the contents of your project files to the server, and asking a human is the only way it can get them.

Once you answer, the task goes back into work and the agent continues with what you said.

Approving dangerous commands

A separate matter from questions. A question is "I don't know, you decide"; an approval is "I know, but I won't without permission".

When an agent is about to run a command that the project rules consider dangerous, it is not executed. Instead an approval request is created, and a human decides: allow or decline.

Requests are visible on the project's "Approvals" screen and on a tab inside the task. A request has a deadline: if nobody answers, it expires, and by default that counts as a refusal, because that is the safer way round.

Which commands count as dangerous is set by the trust profile.

Notifications

So questions and approvals don't sit unnoticed, there are notifications:

• in the platform itself: the notification centre in the top bar, and the Inbox at the top of the sidebar;
• in Telegram, if you have linked a chat;
• in Slack: a project channel for the whole team, and direct messages for what is addressed to you;
• by email, for what concerns you personally: a task assigned to you, an agent waiting for your answer;
• as a device notification, even when the tab is closed;
• in a daily digest email: what happened over the day.

The Inbox collects everything addressed to you across all projects of a workspace. The Needs you tab keeps only what still waits for your move: a question to you or your role without an answer, a command from your agent waiting for approval, work handed to you for review, a task assigned to you that nobody has started. Once it is answered, decided or accepted, it leaves the tab by itself, whoever did it. The number next to Inbox in the sidebar is the size of that tab. All is the whole feed, Mentions is where someone tagged you with @, Done is what you have read and needs nothing more.

Mentions. In a task comment, type @ (or press the @ button under the field) and pick a project member: they get a notification, and the comment lands in their Inbox. You can mention people who can see the project: project members, the owner and administrators. You can't mention an agent: agents don't read the feed as it happens; questions and tasks are how you reach them.

What reaches you and where is set in your profile as a table: a row per event, a column per channel. Under each channel name is a box that turns it on or off for every event at once: Telegram goes quiet with one click, not twenty. Device notifications work wherever you have allowed them in your browser. They are tied to your sign-in on that device: signing out, or revoking that session in the "where you are signed in" list of your profile, turns notifications off on it, and after you sign in again you need to turn them on again. One person can have at most 10 devices; a new one pushes out dead ones first, then the oldest.

Quiet hours

Quiel can leave you alone at night. Turn on quiet hours on the notifications page of your profile and pick the window, 22:00 to 08:00 by default. During it there are no device notifications, no Telegram messages and no Slack direct messages. The bell, the Inbox and email work as usual: a question an agent asks at night does not disappear and waits for you in the morning.

Night messages are not sent later in the morning, otherwise your phone would deliver the whole night at eight. The Inbox and the summary tell you about them instead.

The window follows your time zone, which you set in your profile. If it is not set, turning quiet hours on uses the time zone of the device you turn them on from.

What is in the daily summary

The summary arrives once a day, in the morning in your time zone, and answers two different questions.

What is waiting for you: questions addressed to you or to your role; approvals for your agents; tasks assigned to you and not yet started. This is what is standing still waiting for your action.

What happened in the project: this section goes to workspace owners and admins and to anyone holding the manager role in the project: the same circle that accepts work. What was delivered over the day, with pull request links; what is waiting for acceptance; what is stuck (blocked or waiting for a human); how many tasks are queued per role and how many agents are sitting idle.

A developer does not get the project section: they already see their own tasks, and a list of other people's delivered work is noise.

An empty summary is not sent. If nothing is addressed to you and nothing happened in the project, there is no message.

Project costs

Each task's card shows what it cost: how long it took and how many tokens it consumed. The "Costs" section answers the question you cannot assemble from cards: where the resource goes.

Over a day, a week or a month: how many live tokens were spent, the most expensive tasks with their working time, and a breakdown by agent.

At the top of the page are daily spend for 7, 30 or 90 days (today is highlighted) and the monthly budget. The month is a calendar month in UTC. There are two limits, and each is optional:

• in live tokens, for all agents;
• in money, in the currency of the workspace rates, counting only agents paid per token and only models with a rate. Without rates this limit has no effect: what cannot be counted cannot be called used up.

If both are set, whichever runs out first applies. The budget shows how much is used and when it resets. At 80% and at 100% of each limit the project managers (or the workspace owners, if the project has no manager) get a notification: once per threshold per month, not on every agent turn. Raise the limit and the count starts over.

On its own the budget only warns. If you turn on Pause agents at the limit, then at the limit agents finish the tasks they have started, and new tasks in the project wait for the next month, or until you raise the limit. Whoever can change the project settings can change the budget.

Download CSV gives you the spend for the selected period as a file: one row per day, task, agent and model, with tokens and money. Money follows the same rules as on the page: for subscriptions and for models without a rate the cell is empty, not zero. Export is available on every plan and to the whole project team.

Live tokens means "input + output". Cache reads are shown next to them and smaller, not because they do not matter: they are counted afresh on every request and exceed live tokens by hundreds of times. Add them together and the number starts describing context length rather than work, and the expensive task gets lost behind the verbose one.

What you see on the task itself

Three numbers in the agent block: "input + output", the cache separately, and working time. Next to them, money, when there is something to compute it from: rates are set and the spend came from a metered agent. No rates means no sum, not a zero. If rates cover only some models, what was computed is shown and the rest is stated out loud, otherwise the total would read as complete.

A per-attempt breakdown appears once the task has been sent back from review at least once: "Attempt 1: 220,000, Attempt 2: 110,000". A single total does not answer the question people open the card for: did the rework cost more than the original work or less. The attempt boundaries come from the status feed: there is no second store of the same fact.

By task type

On the Costs screen, next to the per-agent split, sits a split by type: what bugs cost against features. It counts every task in the period, not the ten most expensive: forty cheap bugs never reach that list, and a split computed from the top would say we barely have any bugs.

Task budget

The project settings take a task budget in live tokens. Empty means no budget, and that is the default: a number the platform picked by itself would mark ordinary work as an overrun.

When the spend passes the budget the card gets a mark, the project managers get a notification (once, not on every agent turn), and the agent itself hears about it on delivery and is asked to say in the report where the spend went: the task turned out larger than it looked, or it got stuck going in circles.

Telegram

The bot is linked to your account. In Integrations, Telegram, press "Get a code", then "Open @bot": Telegram opens the bot with the code already in it, and you only press "Start". On a computer the QR code does the same: point your phone camera at it. If the link does not open, send the bot /start CODE by hand; the code works for 15 minutes, once. The screen shows "Linked" on its own, no reload needed.

What the bot can do:

• send notifications about questions and approvals addressed to you;
• answer a question with buttons straight from the chat, if the agent offered options;
• allow and decline dangerous commands with buttons.

An answer from Telegram is the same answer as one in the interface: it lands in the same task and is visible to everyone with access.

Project group or channel

A project feed can go to a shared team chat, a Telegram group or channel, like a Slack channel. It gets: a task was taken or finished, an agent asked a question or is waiting for approval of a dangerous command, a pull request was opened, who allowed or denied a command.

There are no decisions in the group. Under a message there are only "Open" link buttons for tasks that wait for a person. The text of a dangerous command or an agent question never goes to the group; the person named allows the command or answers in the bot private chat or in the interface. Otherwise anyone in the group, including people outside the project, could approve a dangerous command.

How to connect (you need the right to change project settings): project settings, "Telegram group or channel", "Connect a group or channel".

• Group: press "Pick a group in Telegram" (or point your camera at the QR code). Telegram asks you to pick a group and adds the bot; the bot replies that the chat is connected. If the group has topics, send the shown /start@bot CODE command in the topic you want, and the feed goes there.
• Channel: add the bot to the channel as an admin who can post (the link on the screen does it in one tap), then post the shown code in the channel. The bot connects the channel and replies.

Events of one pass are merged into one message: Telegram lets a bot send no more than 20 messages a minute to a group. If the bot is removed from the group, the project disconnects on its own, and whoever connected it and the project managers get a notification. A group that became a supergroup stays connected.

Slack

Slack, like a Telegram group, is a shared surface: the other channels are personal, addressed to one person. Here a project has a channel where the team sees the agents' work without opening the dashboard.

Two roads, and they are deliberately different:

• A project channel: statuses, questions, approvals and opened pull requests, as short lines with a link to the task. Set the channel in the project settings; a project without one works on direct messages alone, and that is a supported mode, not a breakage.
• Direct messages: what is addressed to you personally: an agent's question, a task assigned to you, a dangerous command waiting for your approval. To receive them, link your Slack account: get a code in Integrations and send /quiel link CODE to the bot.

Only someone who can change settings sees the channel list. The channels, private ones the bot was invited to included, are listed only to a person who may change the project or workspace settings, and that person picks the channel. Other members see just the name of the chosen channel in the settings.

The text of a dangerous command never goes to a channel, only the fact that one is waiting, and later who decided. The reason is not tidiness: an approval records who decided, and a button in a shared channel would mean anyone who wandered in could decide.

The app on your computer and phone

You can install Quiel: it opens in its own window, without an address bar or tabs, and sits next to your other apps.

On a computer: in Chrome or Edge, click the install icon at the right of the address bar, or choose “Install Quiel” from the browser menu.

On Android: browser menu, “Install app”.

On iPhone and iPad: Share (the square with an arrow at the bottom of Safari) → “Add to Home Screen”. Then open Quiel from the Home Screen and allow notifications: they do not work in Safari, only in the installed app.

The app is the same Quiel, not a cut-down copy. No separate sign-in is needed.

The task feed

Everything that happened to a task is visible in its feed: status changes, field edits, comments, agent reports, questions and approvals.

The feed answers "who and when", not "word for word": an edit to a long text shows as "added", "removed" or "rewritten", while the text itself is visible in the task.

Comments are written in the same place, in Markdown. The agent sees them. An image from an external address is not loaded by the browser on its own: it is shown as a link, see Security.