feat(ui): Add tool level expansion settings.#563
Conversation
Introduce a central tool-call registry that owns renderer lookup, exact tool names, preset defaults, and the special Other tools entry for unknown tools. Tool and thinking expansion defaults now resolve through the shared preference model instead of hardcoded read/skill behavior. Rework Appearance interaction settings with Minimal, Balanced, Detailed, and Everything presets plus per-item customization rows that switch to Custom when edited. The layout adapts for narrow screens by stacking preset cards and expansion controls. Persist the new toolCallExpansionDefaults setting under the existing UI settings bucket while keeping legacy toolOutputExpansion and thinkingBlocksExpansion synchronized for command-palette compatibility. Validation: npm run typecheck --workspace @codenomad/ui.
|
I want you to perform the gatekeeper checks for latest PR state check for regressions |
|
Gatekeeper review for latest PR state. Assumption: the request is for a strict review only, not code changes. I did not modify the branch. Findings:
This is a real behavioral regression because the rest of the UI still treats
Impact: Required fix: keep Implementation-quality note: The new Validation: I attempted Gatekeeper verdict: request changes before merge. The PR should not ship while -- |
Move normal tool header title resolution onto the renderer that already comes from the central tool registry, eliminating the separate TITLE_RENDERERS map and its stale todoread entry. Keep the task renderer's legacy metadata.summary fallback lightweight by using state titles and generic input details instead of renderer-specific title lookup. This preserves old task summaries without reintroducing a second tool registry. Also remove remaining todoread branches from tool-call utilities because todoread is not an available tool. Validation: npm run typecheck --workspace @codenomad/ui.
|
Addressed the tool-title registry duplication review in follow-up commit f78307d.
Clarification: todoread is not a deprecated tool here; it is not an available tool in this registry/UI, so the stale references were removed rather than kept as a deprecated alias. Validation: npm run typecheck --workspace @codenomad/ui |
|
Now? |
|
PR builds are available as GitHub Actions artifacts: https://github.com/NeuralNomadsAI/CodeNomad/actions/runs/27817555738 Artifacts expire in 7 days.
|
1 similar comment
|
PR builds are available as GitHub Actions artifacts: https://github.com/NeuralNomadsAI/CodeNomad/actions/runs/27817555738 Artifacts expire in 7 days.
|
|
Re-review after follow-up commit Assumption: “Now?” means re-check the latest PR state as gatekeeper, not make code changes. Result: the previous blocking finding is resolved.
Remaining quality concern, not a blocker:
I would not block merge on that because loaded child-session tool calls now render through the real Validation: I re-ran Gatekeeper verdict: no blocking findings in the latest diff. The previous blocker is cleared. Merge risk is acceptable if CI/typecheck is green in the proper dependency environment. -- |
|
PR builds are available as GitHub Actions artifacts: https://github.com/NeuralNomadsAI/CodeNomad/actions/runs/27818237675 Artifacts expire in 7 days.
|
Summary
Validation