Skip to content
Changelog

What's new in Burhan

What changed, week by week, in plain words. Dates are when a change reached every workspace.

  1. Build a journey with Burhan, in the chat

    Tell Ask Burhan what you want tested. It asks, a question or two at a time: what should happen, which test account, whether the flow pays (then a test card) or changes data. It drafts the steps, checks no journey already covers it, and shows the journey for you to confirm, already through the journey review, so Confirm saves it at once. Then a Run now card follows: one more Confirm runs it, and a red or blocked result shows in Attention. Edits Burhan proposes work the same way. Logins, passwords and test cards are typed into the card's own fields, encrypted with the test account, and never go into the chat.

    Attention: reds and blocked runs in one place

    The Blockers page became Attention: everything that needs a person. Reds first, each with Review with Burhan, a ticket and Not a bug; then blocked runs, each with Review with Burhan, Not a blocker and the rest. Rows with a ticket, a Not a bug or a verdict fold under Done. On a phone every row is a card and every menu a bottom sheet. The Journeys page keeps the whole plan: its Failing and Blocked counts now lead here, and the Overview's "See what broke" too. Review with Burhan works on reds as well: it says what broke, whose it looks like, and proposes the ticket or Not a bug.

  2. Ask Burhan: questions and actions on every page

    A panel on the side of every dashboard page. Ask what broke, why a run was blocked, what to fix first or how many credits are left, and Burhan reads the runs and answers. Tell it to run a module, pause a journey, add a doc, change the schedule, invite a teammate or file a ticket, and it proposes the action with a Confirm button; it never pays. On Blockers, Review with Burhan hands it the run and it proposes the verdict and a lesson. Chats are kept per person, across projects, under the clock icon.

    Integrations as apps, and one ticket menu

    Workspace → Integrations shows Slack, ClickUp, GitHub and Jira as tiles with their state. ClickUp is connected once for the workspace, then each project picks its list on the same page. On a red, + Ticket is one menu: create on the connected tracker, connect the other, or copy the issue.

    Credits replace plans

    Burhan is now paid for in credits: 1 credit is $1, a web journey run costs 1 credit and an Android run 2. Every new account starts with $20 of credits, no card. Packs: $20 for 20, $99 for 110, $499 for 600. A run is charged when it finishes; a run that broke on our side costs nothing. Auto-reload buys another pack from your saved card when the balance can't cover a run, up to a monthly cap you set.

    Blockers get a verdict, and Burhan learns from it

    A new Blockers page lists every journey whose last run couldn't finish a step. For each one you say why: Burhan's fault, missing test data, or a real blocker in the app. Burhan's-fault verdicts teach the runner how to do the step next time; test-data verdicts teach it not to write steps that need data your account hasn't got.

    Slack alerts

    Paste an incoming webhook in Workspace → Integrations and Burhan posts to your channel when a run finishes (the tally, credits and time), when a journey goes red (one message per red, with the step), and when credits run low. Pick which of the three you want; send a test message to check the channel.

  3. Scheduled runs start at 09:00, not midnight

    A scheduled full run now starts at 09:00 in your own time zone (Riyadh by default), when shops and branches are open and the results are in by about 10:00. A client's app had rows that only show in opening hours, so midnight runs were blocked for no good reason. Schedules you set yourself are untouched.

    A running full run shows its progress everywhere

    While a full run goes, the header, the project card and the Runs page say how far it is and what it found so far: "Running · 12/36 · 1 failed". A paused run (a login expired) says so, with the button that fixes it.

  4. Android: taps drawn on the video, Android's own dialogs handled

    The run video now shows where Burhan tapped, step by step. Burhan reads Android's own dialogs (permissions, intro slides, a field that already holds a value) and gets past them instead of calling the step blocked. The phone's location is set to where your users are, so "Directions" and "Visit store" land on the right place.

  5. Runs, one row each

    The Runs page lists every run as one row: what it was (the full run, a module, one journey, a retry), when, how long, the tally, and Watch to play the whole run end to end. Journeys and Runs switch at the top of the page. A link to a run opens it already expanded.

  6. The bell: what happened on your projects

    A notification bell in the dashboard: runs that finished, new reds, journeys that went green again, runs waiting for you to log in, journeys Burhan drafted for review, and lessons waiting for your decision. One item per run, never one per journey. Mark all as read when you've seen them.

    What Burhan learned about your app

    A Learned page per project: the lessons Burhan drew from your edits, your "Not a bug" verdicts, the tickets you filed and the rulings you settled. Approve, reject or reword each one. A lesson built only on your own corrections is used from the next run without waiting for us.

  7. A new video player

    The run video has its own player: the steps sit beside it and the video follows the step you pick, taps are drawn on the frame, and the controls work with a keyboard. Phones get the same player at full width.

  8. The dashboard on a phone

    Documents, modules, journeys and runs are laid out for a phone: one row each, a sheet with the actions when you tap, the main button always in reach at the bottom of the screen.

  9. Two new pages on the website

    "AI QA testing" and "Test cases from a PRD", in English and Arabic: what each means and how Burhan does it, with an example journey and the questions people ask.

  10. Twins: a second account of the same kind shares the phones

    On Android, one test account is signed in on one phone at a time (many apps keep one session per account). Write "twin of main" in a second account's description and read-only journeys may run on it in parallel. A journey blocked on the twin (no orders, no point history) runs again on the account it stood in for.

  11. A red stands until a run gets through

    When a journey's newest run couldn't judge (blocked at sign-in, skipped) and the one before it was red, the red is still the answer. A phone that couldn't sign in proves nothing about the bug; a client's real distance bug used to drop out of Failing that way.