| name | zuora-uat-execute-ui |
| description | Internal UI test execution via Playwright MCP |
Execute UI test (internal)
Entry
Prefer starting from hybrid_tr_prepare.py output (called by run-feature after API pytest passes):
python3 "${CLAUDE_PLUGIN_ROOT}/references/uat-test/execution/scripts/hybrid_tr_prepare.py" \
--git-root "$GIT_ROOT" \
--feature "<feature>" \
--tr <n> \
--environment "<environment>"
When execute_ui is true, proceed. Otherwise stop and return the skip_reason.
Manual fallback for debug variables:
python3 "${CLAUDE_PLUGIN_ROOT}/references/uat-test/execution/scripts/parse_debug_variables.py" \
--git-root "$GIT_ROOT" \
--feature "<feature>" \
--tr <n>
Preconditions
Skip UI with a clear message when any of:
hybrid_tr_prepare.py returns execute_ui: false
- Playwright MCP unavailable
- API pytest failed for this TR (
api_test_passed: false)
Process
- Use
variables from hybrid_tr_prepare.py (or parse_debug_variables.py) — do not hand-parse debug logs unless scripts fail
- Read
ui_steps_path / ui_steps_tr{n}.md
- Execute steps via Playwright MCP tools
- Write report to
execution/reports/
- Capture evidence paths for
record-ui-result
Follow ${CLAUDE_PLUGIN_ROOT}/references/uat-test/execution/docs/UI_TEST_EXECUTION_GUIDELINE.md.
MCP
Use user-configured Playwright MCP (not bundled). mcp__zuora-mcp__ask_zuora for unresolved UI behavior only.