Skip to content

test: prune trivial error and layout tests - #8400

Merged
t3-code[bot] merged 1 commit into
mainfrom
test/prune-low-signal-tests-2
Aug 27, 2026
Merged

t3-code[bot] merged 1 commit into
mainfrom
test/prune-low-signal-tests-2

test: prune trivial error and layout tests

746f1fd
Select commit
Loading
Failed to load commit list.
MacroscopeApp / Macroscope - Effect Service Conventions succeeded Aug 27, 2026 in 36s

All clear

All clear

Details

Note

Your check run agent prompt is: .macroscope/check-run-agents/effect-service-conventions.md
More information about how Check Run Agents work can be found in our Help Center.


🔍 Agent activity — 4 tool calls

💭 This PR deletes some test files and un-exports a few symbols—THREAD_SIDEBAR_DEFAULT_WIDTH, initialConfigOption, and CatalogDependencyResolutionError. I need to check whether those symbols are still referenced elsewhere before deciding if removing their export is safe. Deleting tests isn't explicitly called out as a violation, but I should still verify whether removing THREAD_SIDEBAR_DEFAULT_WIDTH's export breaks other consumers, since scripts/ is outside the reviewed scope anyway. Let me grep for usages to confirm nothing else depends on these exports.

1. git_tools — THREAD_SIDEBAR_DEFAULT_WIDTH

2. git_tools — initialConfigOption

💭 Good, no dangling references left. Now I want to check where CatalogDependencyResolutionError is used in the scripts to confirm it's still needed correctly.

3. git_tools — CatalogDependencyResolutionError

4. git_tools — truncate

💭 I don't see broken references, and this PR is really just test pruning plus removing now-unused exports, so no Effect service conventions seem violated. The main thing worth checking is whether deleting error-message tests in apps/server (cliErrors, checkpointing errors, PtyAdapter) conflicts with the convention treating caller-visible error messages as behavior that should be tested. Since this isn't a structural refactor and the messages themselves are unchanged, I don't think that convention applies here. Given everything looks clean, my final response should follow the schema with a success state, since the check only defaults to failure when there are actual findings to report.


Agent Credits: 26 credits