Skip to content

Impact Analysis

When you change a shared keyword, resource, or variable, Impact Analysis answers which tests should you re-run? — not only where the name appears (that is Find References).

  1. Open a Robot file and place the caret on a user keyword, resource import path, or variable (or select the symbol in the outline).
  2. Run Impact Analysis from:
    • Command Palette (⌘⇧P / Ctrl+Shift+P) → Impact Analysis
    • Go → Impact Analysis
  3. A right-hand Impact panel lists affected tests.

Requires an open project with a built analysis graph (open the project and wait for indexing / rebuild if the panel says the graph is empty).

While the panel is open, results refresh after the analysis index updates (for example after you save related files).

Section Meaning
Affected Tests the graph reaches with non-low confidence
Uncertain Low-confidence bindings — review before treating as certain

Each row shows:

  • Test name
  • File, relation (Direct / Transitive / Import / Variable), and confidence
  • A short why line (for example, that the test calls the keyword, imports the resource, or uses the variable)

Click a row to jump to that test — the Impact panel stays open so you can review other hits. Close it with the X.

With Affected results showing, use Run Impact Set (button on the Impact panel, Go → Run Impact Set, or the command palette).

  • Runs only certain (non-low-confidence) affected tests through the normal run pipeline (active run configuration, save-before-run, Stop, reports).
  • Uncertain hits are listed for review but are not included automatically.
  • If there are no certain hits, the action stays disabled / explains that only uncertain hits exist.
  • Large sets use the same Large Run Threshold confirmation as other big runs (Settings → Execution).
  • Prefer Impact when deciding what to re-run after editing a shared keyword, resource, or variable.
  • Use Find References when you need every textual / index occurrence of a name.