Repository navigation
Conversation
Results can explain a failed see or tap on a text or an id without AI. The rules are deterministic and need no network: a failed see or tap on a quoted text gets the line wait until "<text>" appears as its fix and a reason with the time the step ran; an id selector gets a reason only; any other failure text is shown as the reason. Steps that are not see or tap are not covered. The suggestion is carried on RunStep.Triage, so the Results pane and studio_run_state read the same data.
The finished steps of studio_run_state now carry the rule-based triage of a failed selector step. The tool description says so, because agents read the description to know the field is there.
…lector step The Results pane shows the triage of a failed see or tap below its error: a suggested fix line when the rules have one, then the reason. The rows come from triageRows (a pure function, tested in triage-tests) so the rendering rules have a test without a DOM. The frontend suites pass: device-view 116, probescript 11, triage 3. The frontend builds with tsc and vite.
…e MCP table The Unreleased changelog, the Studio Results section and the Studio row of the MCP tool table now say that a failed see or tap carries a rule-based suggested fix and a reason, that it runs without AI, and which steps it does not cover.
The step text quotes both a text and an id, so a consumer cannot tell them apart from the text. StepEvent.SelectorIsID is true for an action step whose selector is an id. The CLI ignores it; the Studio triage uses it to avoid suggesting a text wait for an id.
…eason only The runner writes an id selector in the same quoted form as a text, so the triage parsed an id as a text and suggested wait until "<id>" appears, a wait that would never match. The triage now takes the step event's id flag: an id gets the reason alone, and a text keeps the wait line. Tests cover both, and the RunStep wiring with the flag.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
seeortapon a text or an id shows, under the step in Results and instudio_run_state, a rule-based suggestion: for a text, thewait until "<text>" appearsline to try before the step, plus the reason and the time the step ran.seeortapare not covered.Test plan
go vet ./...andgo test ./...instudio, including the triage testsgo vetandgo test ./internal/mcp/... ./internal/runner/...at the repo root, including the step-event selector testnpm run build,typecheck:device-view,test:device-view(116),test:probescript(11),test:run-scope(10),check:probescript, and the triage tests intriage-tests(3)seeon a text shows await untilsuggestion and a reason under its step🤖 Generated with Claude Code