Qodo

What it does

Qodo looks at a change and writes the tests that are missing, then marks the risk areas and coverage gaps for a reviewer. It is a pre-review step: the output is suggestions, not an approval.

It is most useful on code with few tests and a stable spec. On a codebase where behaviour is undocumented, it will confidently test the wrong thing.

Standout features

• Test generation for changed code
• Coverage gap and risk annotations
• Suggestions accepted one by one
• Works inside the pull request workflow

Main job: surface the tests and risk areas a pull request should have before a human reviews it.

Best for

Use it before review on a change that touches pricing, units or conversion logic, where an untested branch is expensive.

Getting started

Install it on one repository, run it on a small pull request, then read every generated test against the requirement rather than against the code. Accept only the ones that would fail when the behaviour breaks.

Watch out for

Watch out: generated tests often assert what the code currently does, which is not what it should do. Coverage numbers are not proof, and no agent should merge to production unsupervised.