run-qa-pipeline

name	run-qa-pipeline
description	Run the full QA pipeline end-to-end for one feature — plan, generate, then sync. Use when the user says "run the full pipeline for X", "automate end-to-end QA for feature X", or wants one-shot test creation + sync.
argument-hint	<feature-name> [--skip-sync]
allowed-tools	["Read","Write","Edit","Glob","Bash"]

Run QA Pipeline

Run plan-tests → generate-tests → sync-tests in sequence for one feature.

Inputs

• ${user_config.app_name}, ${user_config.profile_path}, ${user_config.target_repo}, ${user_config.output_dir}
• $ARGUMENTS — <feature-name> [--skip-sync]

Steps

1. Validate config. Confirm app_name, profile_path, and (unless --skip-sync) target_repo are set. If any is missing, stop and tell the user which userConfig keys to fill.
2. Run plan-tests with the feature name. Wait for the plan file to be written.
3. Find the plan path that was just written (the newest file in ${user_config.output_dir}/${user_config.app_name}/plans/).
4. Run generate-tests with that plan path.
5. If --skip-sync was not passed, run sync-tests. Recommend --dry-run first if the target repo has uncommitted changes.
6. Report: total elapsed time, files written, any missingSelectors or mocksRequired from generation, sync result. Suggest the next command:

   cd ${user_config.target_repo}
   npx playwright test ${user_config.playwright_test_dir}/<feature-slug>.spec.ts
   

Notes

• The healer is intentionally not part of this pipeline — it requires a failing test + error log as input. After running tests in the target repo, capture failures and invoke /heal-test separately.
• Each stage is also independently invokable; this skill just chains them with shared state.