← ClaudeAtlas

run-testlisted

Run the repo's component tests and report results. Use when the user wants to run tests, or to verify changes before pushing.
qiankunli/devloop · ★ 4 · AI & Automation · score 61
Install: claude install-skill qiankunli/devloop
Run: ``` <PLUGIN_ROOT>/scripts/python <PLUGIN_ROOT>/scripts/run_tests.py [<repo name|path>] [-- <extra args to narrow scope>] ``` Selects components from the explicit target / changed files / clean repo-wide fallback, then runs each component's canonical test command (`make test*`, or `go test ./...` for a Go module without a Makefile target) and stamps `.devloop` validation on success. Skips with guidance if a component has no test command. No `cd` prefix needed — defaults to cwd's repo, then the workspace's last-active repo; pass a subproject name/path to target another. Pass `-- <args>` (e.g. a path or `-k` filter) to narrow scope. Trust the output; report pass/fail. Fix only test code that broke due to the change — never weaken assertions to force a pass. `<PLUGIN_ROOT>` → `${CLAUDE_PLUGIN_ROOT}` on Claude Code; `${PLUGIN_ROOT}` on Codex.