Skip to content

Reading your results

An assessment has four views: Assessment summary, App map, Improvement plan, and History.

The overall assessment shows a score out of 100 and a letter band.

Band Score
A 90 and above
B 80 to 89
C 70 to 79
D 60 to 69
F Below 60

Coverage is the share of supported checks that produced a result. The summary shows it as a percentage, and as the number of supported checks with results out of the number that apply.

A score is only as strong as its coverage. For that reason, no band is shown in two cases:

  • Coverage of supported checks is below 80%. The headline reads Not scored yet.
  • A Critical finding exists. The headline reads Critical issue.

The three groups of findings that matter most, listed first.

Each category shows what needs attention, what is OK, and what was not verified or is outside the scope of this version. See What we check for the categories and the meaning of each outcome.

A finding is a gap the assessment found in the source.

  • Priority is Critical, High, Medium, or Low.
  • Confidence is High, Medium, or Low.
  • Supporting evidence shows the file path and line, a code excerpt, what was observed, and a link to that revision on GitHub where possible.
  • A finding can also list what is still unknown.

For each finding you can ask for an explanation, or mark it as intentional for your team. Marking a finding as intentional records your decision. It does not improve the score.

The plan turns findings into ordered tasks. You can filter by priority and category and change the sort order.

Each task shows:

  • why it matters
  • the recommended action
  • the source evidence
  • how to tell it is done
  • decisions that belong to the owner
  • what it depends on

A task is one of three kinds: a Plan task to fix something, a Verification to confirm something, or an Optional follow-up.

Select Export all, or select tasks and choose Export selected. The plan downloads as a Markdown file that you can give to your coding agent.

A map of the application as the assessment understood it: its surfaces, services, data stores, and outside dependencies.

Assessment history lists every run for an app with its source, revision, score, band, and coverage.

A new assessment of the same GitHub repository is attached to the same app, and your previously confirmed purpose is carried over. To check progress, fix what the plan lists, run a new assessment, and compare the two entries in the history.

You can delete a single assessment from the history. Its source and report data are removed. A record that it existed remains in the workspace activity log.