do-testlisted

# Do Test You are the **test orchestrator**. You parse arguments, dispatch test runners (potentially in parallel), and aggregate results into a summary. ## Variables TEST_ARGS: $ARGUMENTS **If TEST_ARGS is empty or literally `$ARGUMENTS`**: The skill argument substitution did not run. Look at the user's original message in the conversation — they invoked this as `/do-test <argument>`. Extract whatever follows `/do-test` as the value of TEST_ARGS. Do NOT stop or report an error; just use the argument from the message. ## Step 0: Discover Additional Test Skills Before running tests, scan for any additional test-related skill docs in the project: ```bash ls .claude/skills/*test*/*.md .claude/skills-global/*test*/*.md 2>/dev/null ``` **Read any discovered files.** They may define additional test runners, targets, or configurations beyond what this skill covers (e.g., mobile tests, browser tests, performance benchmarks). Incorporate their instructions alongside the defaults below. ## Argument Parsing Parse `TEST_ARGS` to determine what to run: | Input | Behavior | |-------|----------| | _(empty)_ | Run **all** test directories + lint checks | | `unit` | Run `tests/unit/` + lint | | `integration` | Run `tests/integration/` + lint | | `e2e` | Run `tests/e2e/` + lint | | `tools` | Run `tests/tools/` + lint | | `performance` | Run `tests/performance/` + lint | | `tests/unit/test_bridge_logic.py` | Run that specific file + lint | | `--changed` | Detect changed files, map to