Log the bug. Prove the test.
QATestLog is a bug tracker and test-management tool for teams who have to answer one question: is this release ready, and how do we know?
What it does
Bugs that can be reproduced
Steps, expected, actual, environment, screenshots, and the version it was found in. The fields exist because a report missing any of them comes straight back.
Test plans and cases
What is in scope, what is deliberately not, and the cases that decide it. Draft them from your application description and edit what you keep.
Runs that record coverage
Pass, fail, blocked, skipped — and untested. A run that is a third done and all green is not a green release, and the numbers say so.
Teams and roles
Projects belong to a team. Admin, editor or viewer, set per team — a contractor can be a viewer on one project and an editor on another.
A failure becomes a bug
When a case fails, open the bug from the run. It carries the case’s own steps and expected result across, written before anyone knew it would fail.
Duplicates caught early
Before a second report of the same defect reaches a developer, it is compared against what is already open — and you decide, not the machine.
Invite by email
Someone with an account is added and told. Someone without gets a link that creates their account and puts them in the team — no second account, no waiting on an administrator who already has a vouch.
A board for the conversation
Each team gets a message board for the talk around the work — general, or about one project — so a bug’s history stays the record of that defect and nothing else.
ISO/IEC/IEEE 29119 when you need it
Switch a project to the standard and it asks for the sections its document types require, then tells you which are still empty. Per project, because one audited client should not put paperwork on everything else.
Why the untested count is on every screen
Most tools report a pass rate. A pass rate answers a question nobody asked: of the cases somebody happened to run, how many worked. The question a release manager is actually asking is what has not been looked at — and that number is the one that goes missing when a run is closed early on a Friday afternoon.
So every run here starts with its full case list written in as untested, and closing a run with cases unreached writes that count into the record. The tool will let you ship. It will not let you say afterwards that nobody knew.
Where the AI helps, and where it stops
Four things: it turns a rough note into a report with steps a developer can follow, it flags likely duplicates, it drafts a test plan from your description of the application, and it summarises whether a version looks ready. Every one of them lands on a review screen first, with its own confidence next to it and its open questions listed. Nothing reaches the tracker until a person has read it and pressed save, because a QA record nobody read is worse than no record — everyone downstream treats it as though a tester wrote it.
Start with one project
Create an account, make a team, and log the first bug in about two minutes.