Skip to main content
These commands read the problems a fix item covers, and record whether its change fixes each one. A fix item is a work item made from an issue filed for a problem. Its build runs these commands, and you can run them yourself. See Record a fix result. To record a finding, see outerlayer emit finding. Every command on this page takes these options: Each command exits 0 on success. It exits 1, with the reason on stderr, when it refuses or the input is not valid. A command connects the repository to a factory on its own when exactly one factory fits. See Automatic connection.

outerlayer findings show

List the problems a fix item covers, with each problem’s id, its reports and its result. It needs the work.read permission. Keys made with the Agent and Runner presets hold it.
It prints each problem with its title, id, file, reports and result:
With --json, the output is { "problems": [ ... ] }. Each problem has id, title, file, issueKey, result and reports. result is null until a build records one. Then it has result (fixed or not_fixed), note and recordedAt. Each report has findingId, title, file, headSha, itemKey, minutesLost and reportedAt.

outerlayer findings result

Record the build’s result for one problem. It needs the evidence.insert permission. Keys made with the Agent and Runner presets hold it.
<problem> is the problem’s id: the full UUID that findings show prints. Pass exactly one of --fixed and --not-fixed. A second result for the same problem replaces the first. The command refuses a problem that the item does not cover. An item stops covering its problems when its fix merges or its issue closes.
With --json, the output has problemId, result, note and item. A problem counts as fixed only when the build reports it fixed and its pull request merges. See When a problem needs a fix again.