القائمة
Run a manual test of the ralph CLI to verify it works end-to-end. Use when you need to smoke test ralph after making changes.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Run a manual test of the ralph CLI to verify it works end-to-end. Use when you need to smoke test ralph after making changes.
End-to-end ship a feature in this repo — fresh branch from main, local CI gates, commit + push, PR with body, wait for CI, squash-merge, tag at main HEAD, wait for npm publish, verify. Use when the user says "ship this", "release a new version", "create a PR and publish", or describes a code change that should land on npm.
Run CI targets locally, push, and watch remote CI checks. Surfaces issues concisely. All output saved to /tmp.
Enforce test-first workflow for bug fixes. Use whenever the user asks to fix a bug, defect, regression, or issue. Write a failing test that captures the bug AND a passing test for the correct behavior BEFORE touching production code, then fix and flip.
End-to-end ship a feature in this repo — fresh branch from main, local CI gates, commit + push, PR with body, wait for CI, squash-merge, tag at main HEAD, wait for npm publish, verify. Use when the user says "ship this", "release a new version", "create a PR and publish", or describes a code change that should land on npm.
| name | manual-test |
| description | Run a manual test of the ralph CLI to verify it works end-to-end. Use when you need to smoke test ralph after making changes. |
| user-invocable | true |
| argument-hint | ["prompt"] |
Run a manual test of the ralph CLI. Execute these commands sequentially:
rm -rf .ralph/tasks/test
make install
CLAUDECODE="" bun run ralph --name "test" --prompt "$ARGUMENTS"
If no arguments provided, default to: "smoke test - do nothing just skip phases"
.ralph/tasks/test/state.json and report: