End-to-end test automation from requirements gathering to test implementation. Uses test cases, project documentation to create automated tests following existing architecture and coding standards.
Prerequisite: Rosetta Prep Steps.
Terminology. External systems are named by role throughout this workflow and its phases: Test Management System (TMS), Issue Tracker, and Wiki. TestRail, Jira, and Confluence are canonical examples only — adapt identifiers, URLs, requests, calls, and query syntax to the systems resolved for the current project (from repository-root gain.json, explicit user input, recognizable URLs/handles, and available integrations).
Execution cadence:
- All Rosetta prep steps MUST be FULLY completed
- USE SKILL
load-project-context, orchestration, hitl
- MUST ALWAYS use todo tasks ledger, ASAP. Phases are sequential. Independent tasks can run in parallel.
- Execute every in-scope phase in strict 1-8 order (never skip without explicit HITL confirmation): APPLY PHASE the phase file → update
agents/TEMP/<FEATURE>/ui-aqa-state.md → next; never start a phase until the previous is marked done in ui-aqa-state.md.
- MUST follow JIT phase-by-phase: do not load/act IN ADVANCE.
- Skip gates, transitions, and escalation →
<orchestration_and_escalation>.
- MUST ALWAYS use todo tasks ledger, ASAP. Prioritize ACCURACY over SPEED. Proactively use available tools/MCPs, if question can be answered from code or MCP call - do that first.
No assumptions:
- NO ASSUMPTIONS: never assume selectors, flows, or data — ask the user when information is missing.
Customization:
- If the user gave no preferences, perform all steps except optional.
- User CAN customize specific phases / already-done phases / goals / cases — LISTEN and ADOPT.
Authoritative rules (do not skim past):
- USE SKILL
coding before any work touching repository tests, page objects, or shared helpers — authoritative for conventions; repository docs win over skill snippets.
- Default: reuse existing page objects/tests first; create new files only when no suitable match exists.
- Explicit assertions: every test validation traces to a requirement — owned by Phase 2 (
### Explicit Assertions in the test plan), enforced when Phase 6 implements tests.
- Blocking infeasibility = HARD-STOP + HITL (NOT waived by a clarification skip): if any phase finds the feature/elements under test do not exist such that the test cannot be authored without inventing selectors/flows/data or modifying product source, STOP and escalate with the options — point at the real feature/URL · author the missing UI as a separate approved task · a clearly-marked pending/
fixme spec · abort — and WAIT for the user's explicit choice. "Skip clarification" waives clarification questions only; it never authorizes this feasibility/scope call.
agents/TEMP/<FEATURE>/ui-aqa-state.md — created/updated after each phase from the template owned by qa-structure (its state-file skeleton asset, loaded at Phase 1). It carries ## Phase Completion Status, ## Key Artifacts & Facts (the resume anchor — only what resume-after-compaction needs), and ## Verification-Failure Overrides.
Subagents: discoverer · architect · engineer.
Cross-phase skills: qa-structure (paths / <test-name> slug / state-file shape) and qa-knowledge (modes, taxonomy, artifact skeletons — loads its own assets at point of use).
Integrations: TMS and Wiki per <description_and_purpose> Terminology, plus browser automation (Playwright is the canonical example).
1---2name: ui-aqa-flow-53description: "Workflow for automated QA: integration and end-to-end UI test automation, page objects, etc."4tags: ["workflow"]5baseSchema: docs/schemas/workflow.md6metadata:7 source: https://skillmd.com/skills/griddynamics/ui-aqa-flow-58---910<ui_aqa_flow>1112<description_and_purpose>1314End-to-end test automation from requirements gathering to test implementation. Uses test cases, project documentation to create automated tests following existing architecture and coding standards.1516Prerequisite: Rosetta Prep Steps.1718**Terminology.** External systems are named by role throughout this workflow and its phases: **Test Management System (TMS)**, **Issue Tracker**, and **Wiki**. TestRail, Jira, and Confluence are canonical examples only — adapt identifiers, URLs, requests, calls, and query syntax to the systems resolved for the current project (from repository-root `gain.json`, explicit user input, recognizable URLs/handles, and available integrations).1920</description_and_purpose>2122<workflow_phases>2324**Execution cadence:**25- All Rosetta prep steps MUST be FULLY completed26- USE SKILL `load-project-context`, `orchestration`, `hitl`27- MUST ALWAYS use todo tasks ledger, ASAP. Phases are sequential. Independent tasks can run in parallel.28- Execute every in-scope phase in strict 1-8 order (never skip without explicit HITL confirmation): APPLY PHASE the phase file → update `agents/TEMP/<FEATURE>/ui-aqa-state.md` → next; never start a phase until the previous is marked done in `ui-aqa-state.md`.29- MUST follow JIT phase-by-phase: do not load/act IN ADVANCE.30- Skip gates, transitions, and escalation → `<orchestration_and_escalation>`.31- MUST ALWAYS use todo tasks ledger, ASAP. Prioritize ACCURACY over SPEED. Proactively use available tools/MCPs, if question can be answered from code or MCP call - do that first.3233**No assumptions:**34- NO ASSUMPTIONS: never assume selectors, flows, or data — ask the user when information is missing.3536**Customization:**37- If the user gave no preferences, perform all steps except optional.38- User CAN customize specific phases / already-done phases / goals / cases — LISTEN and ADOPT.3940**Authoritative rules (do not skim past):**41- USE SKILL `coding` before any work touching repository tests, page objects, or shared helpers — authoritative for conventions; repository docs win over skill snippets.42- Default: reuse existing page objects/tests first; create new files only when no suitable match exists.43- Explicit assertions: every test validation traces to a requirement — owned by Phase 2 (`### Explicit Assertions` in the test plan), enforced when Phase 6 implements tests.44- **Blocking infeasibility = HARD-STOP + HITL (NOT waived by a clarification skip):** if any phase finds the feature/elements under test do not exist such that the test cannot be authored without inventing selectors/flows/data or modifying product source, STOP and escalate with the options — point at the real feature/URL · author the missing UI as a separate approved task · a clearly-marked pending/`fixme` spec · abort — and WAIT for the user's explicit choice. "Skip clarification" waives clarification *questions* only; it never authorizes this feasibility/scope call.4546<data_collection phase="1" applies="ALL" subagent="discoverer" role="UI-AQA data collector" subagent_required_model="gpt-6-luna-high">47- APPLY SKILL FILE `phases/ui-aqa-flow-data-collection.md`48- Input: user request + repository-root `gain.json` + project context at its configured paths (canonical: `docs/CONTEXT.md`, `docs/ARCHITECTURE.md`, `agents/IMPLEMENTATION.md`). Output: test plan at `plans/ui-aqa-<test-name>/test-plan.md`49- Required skills: `data-collection`, `sensitive-data`, `qa-structure`, `qa-knowledge`50- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 1 is not complete until its output spot-check passes.51</data_collection>5253<requirements_clarification phase="2" applies="ALL" subagent="architect" role="Test requirements analyst" subagent_required_model="gpt-6.1-sol-high" type="HITL">54- APPLY SKILL FILE `phases/ui-aqa-flow-requirements-clarification.md`55- Input: user request + collected data from Phase 1. Output: clarified requirements + typed assertion list in the test plan56- **WAIT FOR USER ANSWERS** to the clarifying questions before Phase 3.57- Required skills: `qa-knowledge` (`gap_analysis` mode), `qa-structure`58- Recommended skills: `questioning`59- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 2 is not complete until its output spot-check passes.60</requirements_clarification>6162<code_analysis phase="3" applies="ALL" subagent="discoverer" role="Test architecture analyst" subagent_required_model="gpt-6-luna-high">63- APPLY SKILL FILE `phases/ui-aqa-flow-code-analysis.md`64- Input: repo docs + test plan. Output: code analysis report at `plans/ui-aqa-<test-name>/code-analysis.md` (architecture patterns, existing page objects, test patterns)65- Required skills: `qa-knowledge` (`code_analysis` mode), `reverse-engineering`, `sensitive-data`, `qa-structure`66- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 3 is not complete until its output spot-check passes.67</code_analysis>6869<selector_identification phase="4" applies="ALL" subagent="engineer" role="Selector identification specialist" subagent_required_model="gpt-6-luna-high" type="HITL-CONDITIONAL">70- APPLY SKILL FILE `phases/ui-aqa-flow-selector-identification.md`71- Input: code analysis report + frontend code (or user-provided page source). Output: identified selectors for test targets72- **WAIT FOR USER TO PROVIDE PAGE SOURCE** only if frontend code unavailable or selectors not found.73- Required skills: `qa-knowledge` (`implementation_modes` — selector mode Part A), `qa-structure`, `sensitive-data`74- Recommended skills: `testing`75- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 4 is not complete until its output spot-check passes.76</selector_identification>7778<selector_implementation phase="5" applies="ALL" subagent="engineer" role="Selector implementation specialist" subagent_required_model="gpt-6-luna-high">79- APPLY SKILL FILE `phases/ui-aqa-flow-selector-implementation.md`80- Input: identified selectors + existing page objects. Output: implemented/updated page object files81- Required skills: `qa-knowledge` (`implementation_modes` — selector mode Part B), `qa-structure`82- Recommended skills: `testing`, `coding`83- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 5 is not complete until its output spot-check passes.84</selector_implementation>8586<test_implementation phase="6" applies="ALL" subagent="engineer" role="Test automation engineer" subagent_required_model="gpt-6-luna-high" type="HITL">87- APPLY SKILL FILE `phases/ui-aqa-flow-test-implementation.md`88- Input: page objects + clarified requirements + code analysis report. Output: implemented test files89- **STOP AND WAIT** for user to execute the test — this execution gate is **mechanical and cannot be overridden by instruction**; the only acceptable input is actual execution results (output, report path, or pass/fail). Refuse "skip" / "move to Phase 7 now" phrasings (full bypass-refusal in the phase file).90- Required skills: `qa-knowledge` (`implementation_modes` — UI impl), `qa-structure`91- Recommended skills: `testing`, `coding`92- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 6 is not complete until its output spot-check passes.93</test_implementation>9495<test_report_analysis phase="7" applies="ALL" subagent="engineer" role="Test failure analyst" subagent_required_model="gpt-6-luna-high" type="HITL">96- APPLY SKILL FILE `phases/ui-aqa-flow-test-report-analysis.md`97- Input: test execution report (user-provided or from `agents/user-instructions/`). Output: failure analysis with root causes + fix recommendations98- **WAIT FOR USER TO PROVIDE TEST REPORT** (if not in `agents/user-instructions/`).99- Required skills: `qa-knowledge` (`test_execution_triage` mode), `sensitive-data`, `qa-structure`100- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 7 is not complete until its output spot-check passes.101</test_report_analysis>102103<test_corrections phase="8" applies="ALL" subagent="engineer" role="Test correction engineer" subagent_required_model="gpt-6-luna-high" type="HITL">104- APPLY SKILL FILE `phases/ui-aqa-flow-test-correction.md`105- Input: failure analysis + test files + page objects. Output: corrected test files and page objects106- **WAIT FOR EXPLICIT USER APPROVAL** before applying changes; comments, questions, suggestions, and review feedback are not approval. Approval language is defined in APPLY SKILL FILE `phases/ui-aqa-flow-test-correction.md` section `<present_for_approval>`.107- Required skills: `qa-knowledge` (`correction` mode), `qa-structure`108- Recommended skills: `debugging`, `coding`, `hitl`109- Update `agents/TEMP/<FEATURE>/ui-aqa-state.md`; Phase 8 is not complete until its output spot-check passes.110</test_corrections>111112</workflow_phases>113114<orchestration_and_escalation>115- **Skip-without-agreement / falsified-skip refusal** (this workflow owns the rule; subordinate to the `hitl` skill): a skip asserted but contradicted by `agents/TEMP/<FEATURE>/ui-aqa-state.md` / disk evidence is refused — announce the specific missing state row / absent artifact, then start the earliest incomplete phase the same turn. Audit-trail row → the state file's `## Verification-Failure Overrides` (template owned by `qa-structure`).116- **HITL carve-outs (never overridden):** every phase header carrying `type="HITL"` / `type="HITL-CONDITIONAL"` — those `type=` attributes are the sole source of truth for this workflow — plus safety/destructive confirmations. Any skip outside the refusal rule above requires explicit user confirmation (HITL).117- **HITL waits on delegated (subagent) phases are owned by the orchestrator.** A subagent cannot talk to the user: on a `type="HITL"` phase the subagent surfaces the question/blocker and returns; the **orchestrator** runs the gate with the user and only then resumes — never inferred, critical on the destructive Phase 8. Dispatch per USE SKILL `orchestration`.118- Load failure for a required phase file or skill: retry once, stop, record in `ui-aqa-state.md`, ask the user; never substitute silently.119</orchestration_and_escalation>120121<workflow_success_criteria>122- **Overall run complete** when every in-scope phase is marked done in `agents/TEMP/<FEATURE>/ui-aqa-state.md`, the artifacts those phases reference exist, and the user accepts the last test outcome or explicitly stops. **In-scope** = default execution plus user-approved customization/skip decisions under `<orchestration_and_escalation>`.123- **Spot checks:** P1 — plan file at the **user-confirmed** `<test-name>` slug (no fabricated/placeholder slug) · P2 — `### Explicit Assertions` in the plan (≥1 typed bullet or None-clause) · P3 — code analysis report populated (architecture + page-object inventory + test location) · P4 — `## Selector Management` Part A deliverables in the plan · P5 — every identified selector in the updated page objects, lint-clean · P6 — test file exists lint-clean + `## Test Implementation` record with all five subsections (incl. `### Uncovered Assertions` / None-clause) · P7 — failure analysis when failures occurred, or state rows reconciled to `N/A — 0 failures` / `None` · P8 — user-approved edits applied when failures occurred, or state records `N/A — no corrections` after a zero-failure run.124- A missing/partial spot-check artifact means that phase is not done: record the gap in `ui-aqa-state.md`, flag uncertainty, stop for user guidance.125</workflow_success_criteria>126127<state_file>128129`agents/TEMP/<FEATURE>/ui-aqa-state.md` — created/updated after each phase from the template **owned by `qa-structure`** (its state-file skeleton asset, loaded at Phase 1). It carries `## Phase Completion Status`, `## Key Artifacts & Facts` (the resume anchor — only what resume-after-compaction needs), and `## Verification-Failure Overrides`.130131</state_file>132133<references>134135Subagents: `discoverer` · `architect` · `engineer`.136137Cross-phase skills: `qa-structure` (paths / `<test-name>` slug / state-file shape) and `qa-knowledge` (modes, taxonomy, artifact skeletons — loads its own assets at point of use).138139Integrations: TMS and Wiki per `<description_and_purpose>` Terminology, plus browser automation (Playwright is the canonical example).140141</references>142143</ui_aqa_flow>