initial commit

This commit is contained in:
2026-03-16 19:54:53 -04:00
commit bfe0e01254
3341 changed files with 483939 additions and 0 deletions

View File

@@ -0,0 +1,226 @@
---
name: 'step-01-preflight-and-context'
description: 'Verify prerequisites and load story, framework, and knowledge base'
outputFile: '{test_artifacts}/atdd-checklist-{story_id}.md'
nextStepFile: './step-02-generation-mode.md'
knowledgeIndex: '{project-root}/_bmad/tea/testarch/tea-index.csv'
---
# Step 1: Preflight & Context Loading
## STEP GOAL
Verify prerequisites and load all required inputs before generating failing tests.
## MANDATORY EXECUTION RULES
- 📖 Read the entire step file before acting
- ✅ Speak in `{communication_language}`
- 🚫 Halt if requirements are missing
---
## EXECUTION PROTOCOLS:
- 🎯 Follow the MANDATORY SEQUENCE exactly
- 💾 Record outputs before proceeding
- 📖 Load the next step only when instructed
## CONTEXT BOUNDARIES:
- Available context: config, loaded artifacts, and knowledge fragments
- Focus: this step's goal only
- Limits: do not execute future steps
- Dependencies: prior steps' outputs (if any)
## MANDATORY SEQUENCE
**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise.
## 1. Stack Detection
**Read `config.test_stack_type`** from `{config_source}`.
**Auto-Detection Algorithm** (when `test_stack_type` is `"auto"` or not configured):
- Scan `{project-root}` for project manifests:
- **Frontend indicators**: `package.json` with react/vue/angular/next dependencies, `playwright.config.*`, `vite.config.*`, `webpack.config.*`
- **Backend indicators**: `pyproject.toml`, `pom.xml`/`build.gradle`, `go.mod`, `*.csproj`/`*.sln`, `Gemfile`, `Cargo.toml`
- **Both present** = `fullstack`; only frontend = `frontend`; only backend = `backend`
- Explicit `test_stack_type` config value overrides auto-detection
- **Backward compatibility**: if `test_stack_type` is not in config, treat as `"auto"` (preserves current frontend behavior for existing installs)
Store result as `{detected_stack}` = `frontend` | `backend` | `fullstack`
---
## 2. Prerequisites (Hard Requirements)
- Story approved with **clear acceptance criteria**
- Test framework configured:
- **If {detected_stack} is `frontend` or `fullstack`:** `playwright.config.ts` or `cypress.config.ts`
- **If {detected_stack} is `backend`:** relevant test config exists (e.g., `conftest.py`, `src/test/`, `*_test.go`, `.rspec`)
- Development environment available
If any are missing: **HALT** and notify the user.
---
## 3. Load Story Context
- Read story markdown from `{story_file}` (or ask user if not provided)
- Extract acceptance criteria and constraints
- Identify affected components and integrations
---
## 4. Load Framework & Existing Patterns
- Read framework config
- Inspect `{test_dir}` for existing test patterns, fixtures, helpers
## 4.5 Read TEA Config Flags
From `{config_source}`:
- `tea_use_playwright_utils`
- `tea_use_pactjs_utils`
- `tea_pact_mcp`
- `tea_browser_automation`
- `test_stack_type`
---
### Tiered Knowledge Loading
Load fragments based on their `tier` classification in `tea-index.csv`:
1. **Core tier** (always load): Foundational fragments required for this workflow
2. **Extended tier** (load on-demand): Load when deeper analysis is needed or when the user's context requires it
3. **Specialized tier** (load only when relevant): Load only when the specific use case matches (e.g., contract-testing only for microservices, email-auth only for email flows)
> **Context Efficiency**: Loading only core fragments reduces context usage by 40-50% compared to loading all fragments.
### Playwright Utils Loading Profiles
**If `tea_use_playwright_utils` is enabled**, select the appropriate loading profile:
- **API-only profile** (when `{detected_stack}` is `backend` or no `page.goto`/`page.locator` found in test files):
Load: `overview`, `api-request`, `auth-session`, `recurse` (~1,800 lines)
- **Full UI+API profile** (when `{detected_stack}` is `frontend`/`fullstack` or browser tests detected):
Load: all Playwright Utils core fragments (~4,500 lines)
**Detection**: Scan `{test_dir}` for files containing `page.goto` or `page.locator`. If none found, use API-only profile.
### Pact.js Utils Loading
**If `tea_use_pactjs_utils` is enabled** (and `{detected_stack}` is `backend` or `fullstack`, or microservices indicators detected):
Load: `pactjs-utils-overview.md`, `pactjs-utils-consumer-helpers.md`, `pactjs-utils-provider-verifier.md`, `pactjs-utils-request-filter.md`
**If `tea_use_pactjs_utils` is disabled** but contract testing is relevant:
Load: `contract-testing.md`
### Pact MCP Loading
**If `tea_pact_mcp` is `"mcp"`:**
Load: `pact-mcp.md`
## 5. Load Knowledge Base Fragments
Use `{knowledgeIndex}` to load:
**Core (always):**
- `data-factories.md`
- `component-tdd.md`
- `test-quality.md`
- `test-healing-patterns.md`
**If {detected_stack} is `frontend` or `fullstack`:**
- `selector-resilience.md`
- `timing-debugging.md`
**Playwright Utils (if enabled and {detected_stack} is `frontend` or `fullstack`):**
- `overview.md`, `api-request.md`, `network-recorder.md`, `auth-session.md`, `intercept-network-call.md`, `recurse.md`, `log.md`, `file-utils.md`, `network-error-monitor.md`, `fixtures-composition.md`
**Playwright CLI (if tea_browser_automation is "cli" or "auto" and {detected_stack} is `frontend` or `fullstack`):**
- `playwright-cli.md`
**MCP Patterns (if tea_browser_automation is "mcp" or "auto" and {detected_stack} is `frontend` or `fullstack`):**
- (existing MCP-related fragments, if any are added in future)
**Traditional Patterns (if utils disabled and {detected_stack} is `frontend` or `fullstack`):**
- `fixture-architecture.md`
- `network-first.md`
**Backend Patterns (if {detected_stack} is `backend` or `fullstack`):**
- `test-levels-framework.md`
- `test-priorities-matrix.md`
- `ci-burn-in.md`
**Pact.js Utils (if enabled):**
- `pactjs-utils-overview.md`, `pactjs-utils-consumer-helpers.md`, `pactjs-utils-provider-verifier.md`, `pactjs-utils-request-filter.md`
**Contract Testing (if pactjs-utils disabled but relevant):**
- `contract-testing.md`
**Pact MCP (if tea_pact_mcp is "mcp"):**
- `pact-mcp.md`
---
## 6. Confirm Inputs
Summarize loaded inputs and confirm with the user. Then proceed.
---
## 7. Save Progress
**Save this step's accumulated work to `{outputFile}`.**
- **If `{outputFile}` does not exist** (first save), create it with YAML frontmatter:
```yaml
---
stepsCompleted: ['step-01-preflight-and-context']
lastStep: 'step-01-preflight-and-context'
lastSaved: '{date}'
---
```
Then write this step's output below the frontmatter.
- **If `{outputFile}` already exists**, update:
- Add `'step-01-preflight-and-context'` to `stepsCompleted` array (only if not already present)
- Set `lastStep: 'step-01-preflight-and-context'`
- Set `lastSaved: '{date}'`
- Append this step's output to the appropriate section.
**Update `inputDocuments`**: Set `inputDocuments` in the output template frontmatter to the list of artifact paths loaded in this step (e.g., knowledge fragments, test design documents, configuration files).
Load next step: `{nextStepFile}`
## 🚨 SYSTEM SUCCESS/FAILURE METRICS:
### ✅ SUCCESS:
- Step completed in full with required outputs
### ❌ SYSTEM FAILURE:
- Skipped sequence steps or missing outputs
**Master Rule:** Skipping steps is FORBIDDEN.

View File

@@ -0,0 +1,96 @@
---
name: 'step-01b-resume'
description: 'Resume interrupted workflow from last completed step'
outputFile: '{test_artifacts}/atdd-checklist-{story_id}.md'
---
# Step 1b: Resume Workflow
## STEP GOAL
Resume an interrupted workflow by loading the existing output document, displaying progress, and routing to the next incomplete step.
## MANDATORY EXECUTION RULES
- 📖 Read the entire step file before acting
- ✅ Speak in `{communication_language}`
---
## EXECUTION PROTOCOLS:
- 🎯 Follow the MANDATORY SEQUENCE exactly
- 📖 Load the next step only when instructed
## CONTEXT BOUNDARIES:
- Available context: Output document with progress frontmatter
- Focus: Load progress and route to next step
- Limits: Do not re-execute completed steps
- Dependencies: Output document must exist from a previous run
## MANDATORY SEQUENCE
**CRITICAL:** Follow this sequence exactly.
### 1. Load Output Document
Read `{outputFile}` and parse YAML frontmatter for:
- `stepsCompleted` — array of completed step names
- `lastStep` — last completed step name
- `lastSaved` — timestamp of last save
**If `{outputFile}` does not exist**, display:
"⚠️ **No previous progress found.** There is no output document to resume from. Please use **[C] Create** to start a fresh workflow run."
**THEN:** Halt. Do not proceed.
---
### 2. Display Progress Dashboard
Display progress with ✅/⬜ indicators:
1. ✅/⬜ Preflight & Context (step-01-preflight-and-context)
2. ✅/⬜ Generation Mode (step-02-generation-mode)
3. ✅/⬜ Test Strategy (step-03-test-strategy)
4. ✅/⬜ Generate Tests + Aggregate (step-04c-aggregate)
5. ✅/⬜ Validate & Complete (step-05-validate-and-complete)
---
### 3. Route to Next Step
Based on `lastStep`, load the next incomplete step:
- `'step-01-preflight-and-context'` → load `./step-02-generation-mode.md`
- `'step-02-generation-mode'` → load `./step-03-test-strategy.md`
- `'step-03-test-strategy'` → load `./step-04-generate-tests.md`
- `'step-04c-aggregate'` → load `./step-05-validate-and-complete.md`
- `'step-05-validate-and-complete'`**Workflow already complete.** Display: "✅ **All steps completed.** Use **[V] Validate** to review outputs or **[E] Edit** to make revisions." Then halt.
**If `lastStep` does not match any value above**, display: "⚠️ **Unknown progress state** (`lastStep`: {lastStep}). Please use **[C] Create** to start fresh." Then halt.
**Otherwise**, load the identified step file, read completely, and execute.
The existing content in `{outputFile}` provides context from previously completed steps.
---
## 🚨 SYSTEM SUCCESS/FAILURE METRICS
### ✅ SUCCESS:
- Output document loaded and parsed correctly
- Progress dashboard displayed accurately
- Routed to correct next step
### ❌ SYSTEM FAILURE:
- Not loading output document
- Incorrect progress display
- Routing to wrong step
**Master Rule:** Resume MUST route to the exact next incomplete step. Never re-execute completed steps.

View File

@@ -0,0 +1,125 @@
---
name: 'step-02-generation-mode'
description: 'Choose AI generation or recording mode'
outputFile: '{test_artifacts}/atdd-checklist-{story_id}.md'
nextStepFile: './step-03-test-strategy.md'
---
# Step 2: Generation Mode Selection
## STEP GOAL
Choose the appropriate generation mode for ATDD tests.
## MANDATORY EXECUTION RULES
- 📖 Read the entire step file before acting
- ✅ Speak in `{communication_language}`
---
## EXECUTION PROTOCOLS:
- 🎯 Follow the MANDATORY SEQUENCE exactly
- 💾 Record outputs before proceeding
- 📖 Load the next step only when instructed
## CONTEXT BOUNDARIES:
- Available context: config, loaded artifacts, and knowledge fragments
- Focus: this step's goal only
- Limits: do not execute future steps
- Dependencies: prior steps' outputs (if any)
## MANDATORY SEQUENCE
**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise.
## 1. Default Mode: AI Generation
Use AI generation when:
- Acceptance criteria are clear
- Scenarios are standard (CRUD, auth, API, navigation)
- **If {detected_stack} is `backend`:** Always use AI generation (no browser recording needed)
Proceed directly to test strategy if this applies.
---
## 2. Optional Mode: Recording (Complex UI)
**Skip this section entirely if {detected_stack} is `backend`.** For backend projects, use AI generation from API documentation, OpenAPI/Swagger specs, or source code analysis instead.
**If {detected_stack} is `frontend` or `fullstack`:**
Use recording when UI interactions need live browser verification.
**Tool selection based on `config.tea_browser_automation`:**
If `auto`:
> **Note:** `${timestamp}` is a placeholder the agent should replace with a unique value (e.g., epoch seconds) for session isolation.
- **Simple recording** (snapshot selectors, capture structure): Use CLI
- `playwright-cli -s=tea-atdd-${timestamp} open <url>``playwright-cli -s=tea-atdd-${timestamp} snapshot` → extract refs
- **Complex recording** (drag/drop, wizards, multi-step state): Use MCP
- Full browser automation with rich tool semantics
- **Fallback:** If preferred tool unavailable, use the other; if neither, skip recording
If `cli`:
- Use Playwright CLI for all recording
- `playwright-cli -s=tea-atdd-${timestamp} open <url>`, `snapshot`, `screenshot`, `click <ref>`, etc.
If `mcp`:
- Use Playwright MCP tools for all recording (current behavior)
- Confirm MCP availability, record selectors and interactions
If `none`:
- Skip recording mode entirely, use AI generation from documentation
---
## 3. Confirm Mode
State the chosen mode and why. Then proceed.
---
## 4. Save Progress
**Save this step's accumulated work to `{outputFile}`.**
- **If `{outputFile}` does not exist** (first save), create it with YAML frontmatter:
```yaml
---
stepsCompleted: ['step-02-generation-mode']
lastStep: 'step-02-generation-mode'
lastSaved: '{date}'
---
```
Then write this step's output below the frontmatter.
- **If `{outputFile}` already exists**, update:
- Add `'step-02-generation-mode'` to `stepsCompleted` array (only if not already present)
- Set `lastStep: 'step-02-generation-mode'`
- Set `lastSaved: '{date}'`
- Append this step's output to the appropriate section.
Load next step: `{nextStepFile}`
## 🚨 SYSTEM SUCCESS/FAILURE METRICS:
### ✅ SUCCESS:
- Step completed in full with required outputs
### ❌ SYSTEM FAILURE:
- Skipped sequence steps or missing outputs
**Master Rule:** Skipping steps is FORBIDDEN.

View File

@@ -0,0 +1,110 @@
---
name: 'step-03-test-strategy'
description: 'Map acceptance criteria to test levels and priorities'
outputFile: '{test_artifacts}/atdd-checklist-{story_id}.md'
nextStepFile: './step-04-generate-tests.md'
---
# Step 3: Test Strategy
## STEP GOAL
Translate acceptance criteria into a prioritized, level-appropriate test plan.
## MANDATORY EXECUTION RULES
- 📖 Read the entire step file before acting
- ✅ Speak in `{communication_language}`
- 🚫 Avoid duplicate coverage across levels
---
## EXECUTION PROTOCOLS:
- 🎯 Follow the MANDATORY SEQUENCE exactly
- 💾 Record outputs before proceeding
- 📖 Load the next step only when instructed
## CONTEXT BOUNDARIES:
- Available context: config, loaded artifacts, and knowledge fragments
- Focus: this step's goal only
- Limits: do not execute future steps
- Dependencies: prior steps' outputs (if any)
## MANDATORY SEQUENCE
**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise.
## 1. Map Acceptance Criteria
- Convert each acceptance criterion into test scenarios
- Include negative and edge cases where risk is high
---
## 2. Select Test Levels
Choose the best level per scenario based on `{detected_stack}`:
**If {detected_stack} is `frontend` or `fullstack`:**
- **E2E** for critical user journeys
- **API** for business logic and service contracts
- **Component** for UI behavior
**If {detected_stack} is `backend` or `fullstack`:**
- **Unit** for pure functions, business logic, and edge cases
- **Integration** for service interactions, database queries, and middleware
- **API/Contract** for endpoint validation, request/response schemas, and Pact contracts
- **No E2E** for pure backend projects (no browser-based testing needed)
---
## 3. Prioritize Tests
Assign P0P3 priorities using risk and business impact.
---
## 4. Confirm Red Phase Requirements
Ensure all tests are designed to **fail before implementation** (TDD red phase).
---
## 5. Save Progress
**Save this step's accumulated work to `{outputFile}`.**
- **If `{outputFile}` does not exist** (first save), create it with YAML frontmatter:
```yaml
---
stepsCompleted: ['step-03-test-strategy']
lastStep: 'step-03-test-strategy'
lastSaved: '{date}'
---
```
Then write this step's output below the frontmatter.
- **If `{outputFile}` already exists**, update:
- Add `'step-03-test-strategy'` to `stepsCompleted` array (only if not already present)
- Set `lastStep: 'step-03-test-strategy'`
- Set `lastSaved: '{date}'`
- Append this step's output to the appropriate section.
Load next step: `{nextStepFile}`
## 🚨 SYSTEM SUCCESS/FAILURE METRICS:
### ✅ SUCCESS:
- Step completed in full with required outputs
### ❌ SYSTEM FAILURE:
- Skipped sequence steps or missing outputs
**Master Rule:** Skipping steps is FORBIDDEN.

View File

@@ -0,0 +1,334 @@
---
name: 'step-04-generate-tests'
description: 'Orchestrate adaptive FAILING test generation (TDD red phase)'
nextStepFile: './step-04c-aggregate.md'
---
# Step 4: Orchestrate Adaptive FAILING Test Generation
## STEP GOAL
Select execution mode deterministically, then generate FAILING API and E2E tests (TDD RED PHASE) with consistent output contracts across agent-team, subagent, or sequential execution.
## MANDATORY EXECUTION RULES
- 📖 Read the entire step file before acting
- ✅ Speak in `{communication_language}`
- ✅ Resolve execution mode from config (`tea_execution_mode`, `tea_capability_probe`)
- ✅ Apply fallback rules deterministically when requested mode is unsupported
- ✅ Generate FAILING tests only (TDD red phase)
- ✅ Wait for required worker steps to complete
- ❌ Do NOT skip capability checks when probing is enabled
- ❌ Do NOT generate passing tests (this is red phase)
- ❌ Do NOT proceed until required worker steps finish
---
## EXECUTION PROTOCOLS:
- 🎯 Follow the MANDATORY SEQUENCE exactly
- 💾 Wait for subagent outputs
- 📖 Load the next step only when instructed
## CONTEXT BOUNDARIES:
- Available context: config, acceptance criteria from Step 1, test strategy from Step 3
- Focus: orchestration only (mode selection + worker dispatch)
- Limits: do not generate tests directly (delegate to worker steps)
- Dependencies: Steps 1-3 outputs
---
## MANDATORY SEQUENCE
**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise.
### 1. Prepare Execution Context
**Generate unique timestamp** for temp file naming:
```javascript
const timestamp = new Date().toISOString().replace(/[:.]/g, '-');
```
**Prepare input context for both subagents:**
```javascript
const parseBooleanFlag = (value, defaultValue = true) => {
if (typeof value === 'string') {
const normalized = value.trim().toLowerCase();
if (['false', '0', 'off', 'no'].includes(normalized)) return false;
if (['true', '1', 'on', 'yes'].includes(normalized)) return true;
}
if (value === undefined || value === null) return defaultValue;
return Boolean(value);
};
const subagentContext = {
story_acceptance_criteria: /* from Step 1 */,
test_strategy: /* from Step 3 */,
knowledge_fragments_loaded: /* list of fragments */,
config: {
test_framework: config.test_framework,
use_playwright_utils: config.tea_use_playwright_utils,
use_pactjs_utils: config.tea_use_pactjs_utils,
pact_mcp: config.tea_pact_mcp, // "mcp" | "none"
browser_automation: config.tea_browser_automation,
execution_mode: config.tea_execution_mode || 'auto', // "auto" | "subagent" | "agent-team" | "sequential"
capability_probe: parseBooleanFlag(config.tea_capability_probe, true), // supports booleans and "false"/"true" strings
provider_endpoint_map: /* from Step 1/3 context, if use_pactjs_utils enabled */,
},
timestamp: timestamp
};
```
---
### 2. Resolve Execution Mode with Capability Probe
```javascript
const normalizeUserExecutionMode = (mode) => {
if (typeof mode !== 'string') return null;
const normalized = mode.trim().toLowerCase().replace(/[-_]/g, ' ').replace(/\s+/g, ' ');
if (normalized === 'auto') return 'auto';
if (normalized === 'sequential') return 'sequential';
if (normalized === 'subagent' || normalized === 'sub agent' || normalized === 'subagents' || normalized === 'sub agents') {
return 'subagent';
}
if (normalized === 'agent team' || normalized === 'agent teams' || normalized === 'agentteam') {
return 'agent-team';
}
return null;
};
const normalizeConfigExecutionMode = (mode) => {
if (mode === 'subagent') return 'subagent';
if (mode === 'auto' || mode === 'sequential' || mode === 'subagent' || mode === 'agent-team') {
return mode;
}
return null;
};
// Explicit user instruction in the active run takes priority over config.
const explicitModeFromUser = normalizeUserExecutionMode(runtime.getExplicitExecutionModeHint?.() || null);
const requestedMode = explicitModeFromUser || normalizeConfigExecutionMode(subagentContext.config.execution_mode) || 'auto';
const probeEnabled = subagentContext.config.capability_probe;
const supports = {
subagent: runtime.canLaunchSubagents?.() === true,
agentTeam: runtime.canLaunchAgentTeams?.() === true,
};
let resolvedMode = requestedMode;
if (requestedMode === 'auto') {
if (supports.agentTeam) resolvedMode = 'agent-team';
else if (supports.subagent) resolvedMode = 'subagent';
else resolvedMode = 'sequential';
} else if (probeEnabled && requestedMode === 'agent-team' && !supports.agentTeam) {
resolvedMode = supports.subagent ? 'subagent' : 'sequential';
} else if (probeEnabled && requestedMode === 'subagent' && !supports.subagent) {
resolvedMode = 'sequential';
}
subagentContext.execution = {
requestedMode,
resolvedMode,
probeEnabled,
supports,
};
if (!probeEnabled && (requestedMode === 'agent-team' || requestedMode === 'subagent')) {
const unsupportedRequestedMode =
(requestedMode === 'agent-team' && !supports.agentTeam) || (requestedMode === 'subagent' && !supports.subagent);
if (unsupportedRequestedMode) {
subagentContext.execution.error = `Requested execution mode "${requestedMode}" is unavailable because capability probing is disabled.`;
throw new Error(subagentContext.execution.error);
}
}
```
Resolution precedence:
1. Explicit user request in this run (`agent team` => `agent-team`; `subagent` => `subagent`; `sequential`; `auto`)
2. `tea_execution_mode` from config
3. Runtime capability fallback (when probing enabled)
If probing is disabled, honor the requested mode strictly. If that mode cannot be executed at runtime, fail with explicit error instead of silent fallback.
---
### 3. Dispatch Worker A: Failing API Test Generation
**Dispatch worker:**
- **Subagent File:** `./step-04a-subagent-api-failing.md`
- **Output File:** `/tmp/tea-atdd-api-tests-${timestamp}.json`
- **Context:** Pass `subagentContext`
- **Execution:**
- `agent-team` or `subagent`: launch non-blocking
- `sequential`: run blocking and wait before next dispatch
- **TDD Phase:** RED (failing tests)
**System Action:**
```
🚀 Launching Subagent A: FAILING API Test Generation (RED PHASE)
📝 Output: /tmp/tea-atdd-api-tests-${timestamp}.json
⚙️ Mode: ${resolvedMode}
🔴 TDD Phase: RED (tests will fail until feature implemented)
⏳ Status: Running...
```
---
### 4. Dispatch Worker B: Failing E2E Test Generation
**Dispatch worker:**
- **Subagent File:** `./step-04b-subagent-e2e-failing.md`
- **Output File:** `/tmp/tea-atdd-e2e-tests-${timestamp}.json`
- **Context:** Pass `subagentContext`
- **Execution:**
- `agent-team` or `subagent`: launch non-blocking
- `sequential`: run blocking and wait before next dispatch
- **TDD Phase:** RED (failing tests)
**System Action:**
```
🚀 Launching Subagent B: FAILING E2E Test Generation (RED PHASE)
📝 Output: /tmp/tea-atdd-e2e-tests-${timestamp}.json
⚙️ Mode: ${resolvedMode}
🔴 TDD Phase: RED (tests will fail until feature implemented)
⏳ Status: Running...
```
---
### 5. Wait for Required Worker Completion
**If `resolvedMode` is `agent-team` or `subagent`:**
```
⏳ Waiting for subagents to complete...
├── Subagent A (API RED): Running... ⟳
└── Subagent B (E2E RED): Running... ⟳
[... time passes ...]
├── Subagent A (API RED): Complete ✅
└── Subagent B (E2E RED): Complete ✅
✅ All subagents completed successfully!
```
**If `resolvedMode` is `sequential`:**
```
✅ Sequential mode: each worker already completed during dispatch.
```
**Verify both outputs exist:**
```javascript
const apiOutputExists = fs.existsSync(`/tmp/tea-atdd-api-tests-${timestamp}.json`);
const e2eOutputExists = fs.existsSync(`/tmp/tea-atdd-e2e-tests-${timestamp}.json`);
if (!apiOutputExists || !e2eOutputExists) {
throw new Error('One or both subagent outputs missing!');
}
```
---
### 6. TDD Red Phase Report
**Display TDD status:**
```
🔴 TDD RED PHASE: Failing Tests Generated
✅ Both subagents completed:
- API Tests: Generated with test.skip()
- E2E Tests: Generated with test.skip()
📋 All tests assert EXPECTED behavior
📋 All tests will FAIL until feature implemented
📋 This is INTENTIONAL (TDD red phase)
Next: Aggregation will verify TDD compliance
```
---
### 7. Execution Report
**Display performance metrics:**
```
🚀 Performance Report:
- Execution Mode: {resolvedMode}
- API Test Generation: ~X minutes
- E2E Test Generation: ~Y minutes
- Total Elapsed: ~mode-dependent
- Parallel Gain: ~50% faster when mode is subagent/agent-team
```
---
### 8. Proceed to Aggregation
**Load aggregation step:**
Load next step: `{nextStepFile}`
The aggregation step (4C) will:
- Read both subagent outputs
- Verify TDD red phase compliance (all tests have test.skip())
- Write all test files to disk
- Generate ATDD checklist
- Calculate summary statistics
---
## EXIT CONDITION
Proceed to Step 4C (Aggregation) when:
- ✅ Subagent A (API failing tests) completed successfully
- ✅ Subagent B (E2E failing tests) completed successfully
- ✅ Both output files exist and are valid JSON
- ✅ TDD red phase status reported
**Do NOT proceed if:**
- ❌ One or both subagents failed
- ❌ Output files missing or corrupted
- ❌ Subagent generated passing tests (wrong - must be failing)
---
## 🚨 SYSTEM SUCCESS/FAILURE METRICS:
### ✅ SUCCESS:
- Both subagents launched successfully
- Both worker steps completed without errors
- Output files generated and valid
- Tests generated with test.skip() (TDD red phase)
- Fallback behavior respected configuration and capability probe rules
### ❌ SYSTEM FAILURE:
- Failed to launch subagents
- One or both subagents failed
- Output files missing or invalid
- Tests generated without test.skip() (wrong phase)
- Unsupported requested mode with probing disabled
**Master Rule:** TDD RED PHASE requires FAILING tests (with test.skip()). Mode selection changes orchestration, never red-phase requirements.

View File

@@ -0,0 +1,286 @@
---
name: 'step-04a-subagent-api-failing'
description: 'Subagent: Generate FAILING API tests (TDD red phase)'
subagent: true
outputFile: '/tmp/tea-atdd-api-tests-{{timestamp}}.json'
---
# Subagent 4A: Generate Failing API Tests (TDD Red Phase)
## SUBAGENT CONTEXT
This is an **isolated subagent** running in parallel with E2E failing test generation.
**What you have from parent workflow:**
- Story acceptance criteria from Step 1
- Test strategy and scenarios from Step 3
- Knowledge fragments loaded: api-request, data-factories, api-testing-patterns
- Config: test framework, Playwright Utils enabled/disabled, Pact.js Utils enabled/disabled (`use_pactjs_utils`), Pact MCP mode (`pact_mcp`)
- Provider Endpoint Map (if `use_pactjs_utils` enabled and provider source accessible)
**Your task:** Generate API tests that will FAIL because the feature is not implemented yet (TDD RED PHASE).
---
## MANDATORY EXECUTION RULES
- 📖 Read this entire subagent file before acting
- ✅ Generate FAILING API tests ONLY
- ✅ Tests MUST fail when run (feature not implemented yet)
- ✅ Output structured JSON to temp file
- ✅ Follow knowledge fragment patterns
- ❌ Do NOT generate E2E tests (that's subagent 4B)
- ❌ Do NOT generate passing tests (this is TDD red phase)
- ❌ Do NOT run tests (that's step 5)
---
## SUBAGENT TASK
### 1. Identify API Endpoints from Acceptance Criteria
From the story acceptance criteria (Step 1 output), identify:
- Which API endpoints will be created for this story
- Expected request/response contracts
- Authentication requirements
- Expected status codes and error scenarios
**Example Acceptance Criteria:**
```
Story: User Registration
- As a user, I can POST to /api/users/register with email and password
- System returns 201 Created with user object
- System returns 400 Bad Request if email already exists
- System returns 422 Unprocessable Entity if validation fails
```
### 2. Generate FAILING API Test Files
For each API endpoint, create test file in `tests/api/[feature].spec.ts`:
**Test Structure (ATDD - Red Phase):**
```typescript
import { test, expect } from '@playwright/test';
// If Playwright Utils enabled:
// import { apiRequest } from '@playwright-utils/api';
test.describe('[Story Name] API Tests (ATDD)', () => {
test.skip('[P0] should register new user successfully', async ({ request }) => {
// THIS TEST WILL FAIL - Endpoint not implemented yet
const response = await request.post('/api/users/register', {
data: {
email: 'newuser@example.com',
password: 'SecurePass123!',
},
});
// Expect 201 but will get 404 (endpoint doesn't exist)
expect(response.status()).toBe(201);
const user = await response.json();
expect(user).toMatchObject({
id: expect.any(Number),
email: 'newuser@example.com',
});
});
test.skip('[P1] should return 400 if email exists', async ({ request }) => {
// THIS TEST WILL FAIL - Endpoint not implemented yet
const response = await request.post('/api/users/register', {
data: {
email: 'existing@example.com',
password: 'SecurePass123!',
},
});
expect(response.status()).toBe(400);
const error = await response.json();
expect(error.message).toContain('Email already exists');
});
});
```
**CRITICAL ATDD Requirements:**
- ✅ Use `test.skip()` to mark tests as intentionally failing (red phase)
- ✅ Write assertions for EXPECTED behavior (even though not implemented)
- ✅ Use realistic test data (not placeholder data)
- ✅ Test both happy path and error scenarios from acceptance criteria
- ✅ Use `apiRequest()` helper if Playwright Utils enabled
- ✅ Use data factories for test data (from data-factories fragment)
- ✅ Include priority tags [P0], [P1], [P2], [P3]
### 1.5 Provider Source Scrutiny for CDC in TDD Red Phase (If `use_pactjs_utils` Enabled)
When generating Pact consumer contract tests in the ATDD red phase, provider scrutiny applies with TDD-specific rules. Apply the **Seven-Point Scrutiny Checklist** from `contract-testing.md` (Response shape, Status codes, Field names, Enum values, Required fields, Data types, Nested structures) for both existing and new endpoints.
**If provider endpoint already exists** (extending an existing API):
- READ the provider route handler, types, and validation schemas
- Verify all seven scrutiny points against the provider source: Response shape, Status codes, Field names, Enum values, Required fields, Data types, Nested structures
- Add `// Provider endpoint:` comment and scrutiny evidence block documenting findings for each point
- Wrap the entire test function in `test.skip()` (so the whole test including `executeTest` is skipped), not just the callback
**If provider endpoint is new** (TDD — endpoint not implemented yet):
- Use acceptance criteria as the source of truth for expected behavior
- Acceptance criteria should specify all seven scrutiny points where possible (status codes, field names, types, etc.) — note any gaps as assumptions in the evidence block
- Add `// Provider endpoint: TODO — new endpoint, not yet implemented`
- Document expected behavior from acceptance criteria in scrutiny evidence block
- Wrap the entire test function in `test.skip()` and use realistic expectations from the story
**Graceful degradation when provider source is inaccessible:**
1. **OpenAPI/Swagger spec available**: Use the spec as the source of truth for response shapes, status codes, and field names
2. **Pact Broker available** (when `pact_mcp` is `"mcp"`): Use SmartBear MCP tools to fetch existing provider states and verified interactions as reference
3. **Neither available**: For new endpoints, use acceptance criteria; for existing endpoints, use consumer-side types. Mark with `// Provider endpoint: TODO — provider source not accessible, verify manually` and set `provider_scrutiny: "pending"` in output JSON
4. **Never silently guess**: Document all assumptions in the scrutiny evidence block
**Provider endpoint comments are MANDATORY** even in red-phase tests — they document the intent.
**Example: Red-phase Pact test with provider scrutiny:**
```typescript
// Provider endpoint: TODO — new endpoint, not yet implemented
/*
* Provider Scrutiny Evidence:
* - Handler: NEW — not yet implemented (TDD red phase)
* - Expected from acceptance criteria:
* - Endpoint: POST /api/v2/users/register
* - Status: 201 for success, 400 for duplicate email, 422 for validation error
* - Response: { id: number, email: string, createdAt: string }
*/
test.skip('[P0] should generate consumer contract for user registration', async () => {
await provider
.given('no users exist')
.uponReceiving('a request to register a new user')
.withRequest({
method: 'POST',
path: '/api/v2/users/register',
headers: { 'Content-Type': 'application/json' },
body: { email: 'newuser@example.com', password: 'SecurePass123!' },
})
.willRespondWith({
status: 201,
headers: { 'Content-Type': 'application/json' },
body: like({
id: integer(1),
email: string('newuser@example.com'),
createdAt: string('2025-01-15T10:00:00Z'),
}),
})
.executeTest(async (mockServer) => {
const result = await registerUser({ email: 'newuser@example.com', password: 'SecurePass123!' }, { baseUrl: mockServer.url });
expect(result.id).toEqual(expect.any(Number));
});
});
```
**Why test.skip():**
- Tests are written correctly for EXPECTED behavior
- But we know they'll fail because feature isn't implemented
- `test.skip()` documents this is intentional (TDD red phase)
- Once feature is implemented, remove `test.skip()` to verify green phase
### 3. Track Fixture Needs
Identify fixtures needed for API tests:
- Authentication fixtures (if endpoints require auth)
- Data factories (user data, etc.)
- API client configurations
**Do NOT create fixtures yet** - just track what's needed for aggregation step.
---
## OUTPUT FORMAT
Write JSON to temp file: `/tmp/tea-atdd-api-tests-{{timestamp}}.json`
```json
{
"success": true,
"subagent": "atdd-api-tests",
"tests": [
{
"file": "tests/api/user-registration.spec.ts",
"content": "[full TypeScript test file content with test.skip()]",
"description": "ATDD API tests for user registration (RED PHASE)",
"expected_to_fail": true,
"acceptance_criteria_covered": [
"User can register with email/password",
"System returns 201 on success",
"System returns 400 if email exists"
],
"priority_coverage": {
"P0": 1,
"P1": 2,
"P2": 0,
"P3": 0
}
}
],
"fixture_needs": ["userDataFactory"],
"knowledge_fragments_used": ["api-request", "data-factories", "api-testing-patterns"],
"test_count": 3,
"tdd_phase": "RED",
"provider_scrutiny": "completed",
"summary": "Generated 3 FAILING API tests for user registration story"
}
```
**On Error:**
```json
{
"success": false,
"subagent": "atdd-api-tests",
"error": "Error message describing what went wrong",
"partial_output": {
/* any tests generated before error */
}
}
```
---
## EXIT CONDITION
Subagent completes when:
- ✅ All API endpoints from acceptance criteria have test files
- ✅ All tests use `test.skip()` (documented failing tests)
- ✅ All tests assert EXPECTED behavior (not placeholder assertions)
- ✅ JSON output written to temp file
- ✅ Fixture needs to be tracked
**Subagent terminates here.** Parent workflow will read output and proceed to aggregation.
---
## 🚨 SUBAGENT SUCCESS METRICS
### ✅ SUCCESS:
- All API tests generated with test.skip()
- Tests assert expected behavior (not placeholders)
- JSON output valid and complete
- No E2E/component/unit tests included (out of scope)
- Tests follow knowledge fragment patterns
- Every Pact interaction has `// Provider endpoint:` comment (if CDC enabled)
- Provider scrutiny completed or TODO markers added for new endpoints (if CDC enabled)
### ❌ FAILURE:
- Generated passing tests (wrong - this is RED phase)
- Tests without test.skip() (will break CI)
- Placeholder assertions (expect(true).toBe(true))
- Did not follow knowledge fragment patterns
- Invalid or missing JSON output
- Pact interactions missing provider endpoint comments (if CDC enabled)

View File

@@ -0,0 +1,244 @@
---
name: 'step-04b-subagent-e2e-failing'
description: 'Subagent: Generate FAILING E2E tests (TDD red phase)'
subagent: true
outputFile: '/tmp/tea-atdd-e2e-tests-{{timestamp}}.json'
---
# Subagent 4B: Generate Failing E2E Tests (TDD Red Phase)
## SUBAGENT CONTEXT
This is an **isolated subagent** running in parallel with API failing test generation.
**What you have from parent workflow:**
- Story acceptance criteria from Step 1
- Test strategy and user journey scenarios from Step 3
- Knowledge fragments loaded: fixture-architecture, network-first, selector-resilience
- Config: test framework, Playwright Utils enabled/disabled
**Your task:** Generate E2E tests that will FAIL because the feature UI is not implemented yet (TDD RED PHASE).
---
## MANDATORY EXECUTION RULES
- 📖 Read this entire subagent file before acting
- ✅ Generate FAILING E2E tests ONLY
- ✅ Tests MUST fail when run (UI not implemented yet)
- ✅ Output structured JSON to temp file
- ✅ Follow knowledge fragment patterns
- ❌ Do NOT generate API tests (that's subagent 4A)
- ❌ Do NOT generate passing tests (this is TDD red phase)
- ❌ Do NOT run tests (that's step 5)
---
## SUBAGENT TASK
### 1. Identify User Journeys from Acceptance Criteria
From the story acceptance criteria (Step 1 output), identify:
- Which UI flows will be created for this story
- User interactions required
- Expected visual states
- Success/error messages expected
**Example Acceptance Criteria:**
```
Story: User Registration
- As a user, I can navigate to /register page
- I can fill in email and password fields
- I can click "Register" button
- System shows success message and redirects to dashboard
- System shows error if email already exists
```
### 2. Browser Interaction (Selector Verification)
**Automation mode:** `config.tea_browser_automation`
If `auto` (fall back to MCP if CLI unavailable; if neither available, generate from best practices):
- Open the target page first, then verify selectors with a snapshot:
`playwright-cli -s=tea-atdd-{{timestamp}} open <target_url>`
`playwright-cli -s=tea-atdd-{{timestamp}} snapshot` → map refs to Playwright locators
- ref `{role: "button", name: "Submit"}``page.getByRole('button', { name: 'Submit' })`
- ref `{role: "textbox", name: "Email"}``page.getByRole('textbox', { name: 'Email' })`
- `playwright-cli -s=tea-atdd-{{timestamp}} close` when done
If `cli` (CLI only — do NOT fall back to MCP; generate from best practices if CLI unavailable):
- Open the target page first, then verify selectors with a snapshot:
`playwright-cli -s=tea-atdd-{{timestamp}} open <target_url>`
`playwright-cli -s=tea-atdd-{{timestamp}} snapshot` → map refs to Playwright locators
- ref `{role: "button", name: "Submit"}``page.getByRole('button', { name: 'Submit' })`
- ref `{role: "textbox", name: "Email"}``page.getByRole('textbox', { name: 'Email' })`
- `playwright-cli -s=tea-atdd-{{timestamp}} close` when done
> **Session Hygiene:** Always close sessions using `playwright-cli -s=tea-atdd-{{timestamp}} close`. Do NOT use `close-all` — it kills every session on the machine and breaks parallel execution.
If `mcp`:
- Use MCP tools for selector verification (current behavior)
If `none`:
- Generate selectors from best practices without browser verification
### 3. Generate FAILING E2E Test Files
For each user journey, create test file in `tests/e2e/[feature].spec.ts`:
**Test Structure (ATDD - Red Phase):**
```typescript
import { test, expect } from '@playwright/test';
test.describe('[Story Name] E2E User Journey (ATDD)', () => {
test.skip('[P0] should complete user registration successfully', async ({ page }) => {
// THIS TEST WILL FAIL - UI not implemented yet
await page.goto('/register');
// Expect registration form but will get 404 or missing elements
await page.fill('[name="email"]', 'newuser@example.com');
await page.fill('[name="password"]', 'SecurePass123!');
await page.click('button:has-text("Register")');
// Expect success message and redirect
await expect(page.getByText('Registration successful!')).toBeVisible();
await page.waitForURL('/dashboard');
});
test.skip('[P1] should show error if email exists', async ({ page }) => {
// THIS TEST WILL FAIL - UI not implemented yet
await page.goto('/register');
await page.fill('[name="email"]', 'existing@example.com');
await page.fill('[name="password"]', 'SecurePass123!');
await page.click('button:has-text("Register")');
// Expect error message
await expect(page.getByText('Email already exists')).toBeVisible();
});
});
```
**CRITICAL ATDD Requirements:**
- ✅ Use `test.skip()` to mark tests as intentionally failing (red phase)
- ✅ Write assertions for EXPECTED UI behavior (even though not implemented)
- ✅ Use resilient selectors: getByRole, getByText, getByLabel (from selector-resilience)
- ✅ Follow network-first patterns if API calls involved (from network-first)
- ✅ Test complete user journeys from acceptance criteria
- ✅ Include priority tags [P0], [P1], [P2], [P3]
- ✅ Use proper TypeScript types
- ✅ Deterministic waits (no hard sleeps)
**Why test.skip():**
- Tests are written correctly for EXPECTED UI behavior
- But we know they'll fail because UI isn't implemented
- `test.skip()` documents this is intentional (TDD red phase)
- Once UI is implemented, remove `test.skip()` to verify green phase
### 4. Track Fixture Needs
Identify fixtures needed for E2E tests:
- Authentication fixtures (if journey requires logged-in state)
- Network mocks (if API calls involved)
- Test data fixtures
**Do NOT create fixtures yet** - just track what's needed for aggregation step.
---
## OUTPUT FORMAT
Write JSON to temp file: `/tmp/tea-atdd-e2e-tests-{{timestamp}}.json`
```json
{
"success": true,
"subagent": "atdd-e2e-tests",
"tests": [
{
"file": "tests/e2e/user-registration.spec.ts",
"content": "[full TypeScript test file content with test.skip()]",
"description": "ATDD E2E tests for user registration journey (RED PHASE)",
"expected_to_fail": true,
"acceptance_criteria_covered": [
"User can navigate to /register",
"User can fill registration form",
"System shows success message on registration",
"System shows error if email exists"
],
"priority_coverage": {
"P0": 1,
"P1": 1,
"P2": 0,
"P3": 0
}
}
],
"fixture_needs": ["registrationPageMock"],
"knowledge_fragments_used": ["fixture-architecture", "network-first", "selector-resilience"],
"test_count": 2,
"tdd_phase": "RED",
"summary": "Generated 2 FAILING E2E tests for user registration story"
}
```
**On Error:**
```json
{
"success": false,
"subagent": "atdd-e2e-tests",
"error": "Error message describing what went wrong",
"partial_output": {
/* any tests generated before error */
}
}
```
---
## EXIT CONDITION
Subagent completes when:
- ✅ All user journeys from acceptance criteria have test files
- ✅ All tests use `test.skip()` (documented failing tests)
- ✅ All tests assert EXPECTED UI behavior (not placeholder assertions)
- ✅ Resilient selectors used (getByRole, getByText)
- ✅ JSON output written to temp file
- ✅ Fixture needs tracked
**Subagent terminates here.** Parent workflow will read output and proceed to aggregation.
---
## 🚨 SUBAGENT SUCCESS METRICS
### ✅ SUCCESS:
- All E2E tests generated with test.skip()
- Tests assert expected UI behavior (not placeholders)
- Resilient selectors used (getByRole, getByText)
- JSON output valid and complete
- No API/component/unit tests included (out of scope)
- Tests follow knowledge fragment patterns
### ❌ FAILURE:
- Generated passing tests (wrong - this is RED phase)
- Tests without test.skip() (will break CI)
- Placeholder assertions (expect(true).toBe(true))
- Brittle selectors used (CSS classes, XPath)
- Did not follow knowledge fragment patterns
- Invalid or missing JSON output

View File

@@ -0,0 +1,370 @@
---
name: 'step-04c-aggregate'
description: 'Aggregate subagent outputs and complete ATDD test infrastructure'
outputFile: '{test_artifacts}/atdd-checklist-{story_id}.md'
nextStepFile: './step-05-validate-and-complete.md'
---
# Step 4C: Aggregate ATDD Test Generation Results
## STEP GOAL
Read outputs from parallel subagents (API + E2E failing test generation), aggregate results, verify TDD red phase compliance, and create supporting infrastructure.
---
## MANDATORY EXECUTION RULES
- 📖 Read the entire step file before acting
- ✅ Speak in `{communication_language}`
- ✅ Read subagent outputs from temp files
- ✅ Verify all tests are marked with test.skip() (TDD red phase)
- ✅ Generate shared fixtures based on fixture needs
- ✅ Write all generated test files to disk
- ❌ Do NOT remove test.skip() (that's done after feature implementation)
- ❌ Do NOT run tests yet (that's step 5 - verify they fail)
---
## EXECUTION PROTOCOLS:
- 🎯 Follow the MANDATORY SEQUENCE exactly
- 💾 Record outputs before proceeding
- 📖 Load the next step only when instructed
## CONTEXT BOUNDARIES:
- Available context: config, subagent outputs from temp files
- Focus: aggregation and TDD validation
- Limits: do not execute future steps
- Dependencies: Step 4A and 4B subagent outputs
---
## MANDATORY SEQUENCE
**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise.
### 1. Read Subagent Outputs
**Read API test subagent output:**
```javascript
const apiTestsPath = '/tmp/tea-atdd-api-tests-{{timestamp}}.json';
const apiTestsOutput = JSON.parse(fs.readFileSync(apiTestsPath, 'utf8'));
```
**Read E2E test subagent output:**
```javascript
const e2eTestsPath = '/tmp/tea-atdd-e2e-tests-{{timestamp}}.json';
const e2eTestsOutput = JSON.parse(fs.readFileSync(e2eTestsPath, 'utf8'));
```
**Verify both subagents succeeded:**
- Check `apiTestsOutput.success === true`
- Check `e2eTestsOutput.success === true`
- If either failed, report error and stop (don't proceed)
---
### 2. Verify TDD Red Phase Compliance
**CRITICAL TDD Validation:**
**Check API tests:**
```javascript
apiTestsOutput.tests.forEach((test) => {
// Verify test.skip() is present
if (!test.content.includes('test.skip(')) {
throw new Error(`ATDD ERROR: ${test.file} missing test.skip() - tests MUST be skipped in red phase!`);
}
// Verify not placeholder assertions
if (test.content.includes('expect(true).toBe(true)')) {
throw new Error(`ATDD ERROR: ${test.file} has placeholder assertions - must assert EXPECTED behavior!`);
}
// Verify expected_to_fail flag
if (!test.expected_to_fail) {
throw new Error(`ATDD ERROR: ${test.file} not marked as expected_to_fail!`);
}
});
```
**Check E2E tests:**
```javascript
e2eTestsOutput.tests.forEach((test) => {
// Same validation as API tests
if (!test.content.includes('test.skip(')) {
throw new Error(`ATDD ERROR: ${test.file} missing test.skip() - tests MUST be skipped in red phase!`);
}
if (test.content.includes('expect(true).toBe(true)')) {
throw new Error(`ATDD ERROR: ${test.file} has placeholder assertions!`);
}
if (!test.expected_to_fail) {
throw new Error(`ATDD ERROR: ${test.file} not marked as expected_to_fail!`);
}
});
```
**If validation passes:**
```
✅ TDD Red Phase Validation: PASS
- All tests use test.skip()
- All tests assert expected behavior (not placeholders)
- All tests marked as expected_to_fail
```
---
### 3. Write All Test Files to Disk
**Write API test files:**
```javascript
apiTestsOutput.tests.forEach((test) => {
fs.writeFileSync(test.file, test.content, 'utf8');
console.log(`✅ Created (RED): ${test.file}`);
});
```
**Write E2E test files:**
```javascript
e2eTestsOutput.tests.forEach((test) => {
fs.writeFileSync(test.file, test.content, 'utf8');
console.log(`✅ Created (RED): ${test.file}`);
});
```
---
### 4. Aggregate Fixture Needs
**Collect all fixture needs from both subagents:**
```javascript
const allFixtureNeeds = [...apiTestsOutput.fixture_needs, ...e2eTestsOutput.fixture_needs];
// Remove duplicates
const uniqueFixtures = [...new Set(allFixtureNeeds)];
```
---
### 5. Generate Fixture Infrastructure
**Create fixtures needed by ATDD tests:**
(Similar to automate workflow, but may be simpler for ATDD since feature not implemented)
**Minimal fixtures for TDD red phase:**
```typescript
// tests/fixtures/test-data.ts
export const testUserData = {
email: 'test@example.com',
password: 'SecurePass123!',
};
```
Note: More complete fixtures will be needed when moving to green phase.
---
### 6. Generate ATDD Checklist
**Create ATDD checklist document:**
```markdown
# ATDD Checklist: [Story Name]
## TDD Red Phase (Current)
✅ Failing tests generated
- API Tests: {api_test_count} tests (all skipped)
- E2E Tests: {e2e_test_count} tests (all skipped)
## Acceptance Criteria Coverage
{list all acceptance criteria with test coverage}
## Next Steps (TDD Green Phase)
After implementing the feature:
1. Remove `test.skip()` from all test files
2. Run tests: `npm test`
3. Verify tests PASS (green phase)
4. If any tests fail:
- Either fix implementation (feature bug)
- Or fix test (test bug)
5. Commit passing tests
## Implementation Guidance
Feature endpoints to implement:
{list endpoints from API tests}
UI components to implement:
{list UI flows from E2E tests}
```
**Save checklist:**
```javascript
fs.writeFileSync(`{test_artifacts}/atdd-checklist-{story-id}.md`, checklistContent, 'utf8');
```
---
### 7. Calculate Summary Statistics
**Aggregate test counts:**
```javascript
const resolvedMode = subagentContext?.execution?.resolvedMode; // Provided by Step 4's orchestration context
const subagentExecutionLabel =
resolvedMode === 'sequential'
? 'SEQUENTIAL (API → E2E)'
: resolvedMode === 'agent-team'
? 'AGENT-TEAM (API + E2E)'
: resolvedMode === 'subagent'
? 'SUBAGENT (API + E2E)'
: 'PARALLEL (API + E2E)';
const performanceGainLabel =
resolvedMode === 'sequential'
? 'baseline (no parallel speedup)'
: resolvedMode === 'agent-team' || resolvedMode === 'subagent'
? '~50% faster than sequential'
: 'mode-dependent';
const summary = {
tdd_phase: 'RED',
total_tests: apiTestsOutput.test_count + e2eTestsOutput.test_count,
api_tests: apiTestsOutput.test_count,
e2e_tests: e2eTestsOutput.test_count,
all_tests_skipped: true,
expected_to_fail: true,
fixtures_created: uniqueFixtures.length,
acceptance_criteria_covered: [
...apiTestsOutput.tests.flatMap((t) => t.acceptance_criteria_covered),
...e2eTestsOutput.tests.flatMap((t) => t.acceptance_criteria_covered),
],
knowledge_fragments_used: [...apiTestsOutput.knowledge_fragments_used, ...e2eTestsOutput.knowledge_fragments_used],
subagent_execution: subagentExecutionLabel,
performance_gain: performanceGainLabel,
};
```
**Store summary for Step 5:**
```javascript
fs.writeFileSync('/tmp/tea-atdd-summary-{{timestamp}}.json', JSON.stringify(summary, null, 2), 'utf8');
```
---
## OUTPUT SUMMARY
Display to user:
```
✅ ATDD Test Generation Complete (TDD RED PHASE)
🔴 TDD Red Phase: Failing Tests Generated
📊 Summary:
- Total Tests: {total_tests} (all with test.skip())
- API Tests: {api_tests} (RED)
- E2E Tests: {e2e_tests} (RED)
- Fixtures Created: {fixtures_created}
- All tests will FAIL until feature implemented
✅ Acceptance Criteria Coverage:
{list all covered criteria}
🚀 Performance: {performance_gain}
📂 Generated Files:
- tests/api/[feature].spec.ts (with test.skip())
- tests/e2e/[feature].spec.ts (with test.skip())
- tests/fixtures/test-data.ts
- {test_artifacts}/atdd-checklist-{story-id}.md
📝 Next Steps:
1. Implement the feature
2. Remove test.skip() from tests
3. Run tests → verify PASS (green phase)
4. Commit passing tests
✅ Ready for validation (Step 5 - verify tests fail as expected)
```
---
## EXIT CONDITION
Proceed to Step 5 when:
- ✅ All test files written to disk (API + E2E)
- ✅ All tests verified to have test.skip()
- ✅ All fixtures created
- ✅ ATDD checklist generated
- ✅ Summary statistics calculated and saved
- ✅ Output displayed to user
---
### 8. Save Progress
**Save this step's accumulated work to `{outputFile}`.**
- **If `{outputFile}` does not exist** (first save), create it with YAML frontmatter:
```yaml
---
stepsCompleted: ['step-04c-aggregate']
lastStep: 'step-04c-aggregate'
lastSaved: '{date}'
---
```
Then write this step's output below the frontmatter.
- **If `{outputFile}` already exists**, update:
- Add `'step-04c-aggregate'` to `stepsCompleted` array (only if not already present)
- Set `lastStep: 'step-04c-aggregate'`
- Set `lastSaved: '{date}'`
- Append this step's output to the appropriate section.
Load next step: `{nextStepFile}`
---
## 🚨 SYSTEM SUCCESS/FAILURE METRICS:
### ✅ SUCCESS:
- Both subagents succeeded
- All tests have test.skip() (TDD red phase compliant)
- All tests assert expected behavior (not placeholders)
- All test files written to disk
- ATDD checklist generated
### ❌ SYSTEM FAILURE:
- One or both subagents failed
- Tests missing test.skip() (would break CI)
- Tests have placeholder assertions
- Test files not written to disk
- ATDD checklist missing
**Master Rule:** TDD RED PHASE requires ALL tests to use test.skip() and assert expected behavior.

View File

@@ -0,0 +1,106 @@
---
name: 'step-05-validate-and-complete'
description: 'Validate ATDD outputs and summarize'
outputFile: '{test_artifacts}/atdd-checklist-{story_id}.md'
---
# Step 5: Validate & Complete
## STEP GOAL
Validate ATDD outputs and provide a completion summary.
## MANDATORY EXECUTION RULES
- 📖 Read the entire step file before acting
- ✅ Speak in `{communication_language}`
- ✅ Validate against the checklist
---
## EXECUTION PROTOCOLS:
- 🎯 Follow the MANDATORY SEQUENCE exactly
- 💾 Record outputs before proceeding
- 📖 Load the next step only when instructed
## CONTEXT BOUNDARIES:
- Available context: config, loaded artifacts, and knowledge fragments
- Focus: this step's goal only
- Limits: do not execute future steps
- Dependencies: prior steps' outputs (if any)
## MANDATORY SEQUENCE
**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise.
## 1. Validation
Use `checklist.md` to validate:
- Prerequisites satisfied
- Test files created correctly
- Checklist matches acceptance criteria
- Tests are designed to fail before implementation
- [ ] CLI sessions cleaned up (no orphaned browsers)
- [ ] Temp artifacts stored in `{test_artifacts}/` not random locations
Fix any gaps before completion.
---
## 2. Polish Output
Before finalizing, review the complete output document for quality:
1. **Remove duplication**: Progressive-append workflow may have created repeated sections — consolidate
2. **Verify consistency**: Ensure terminology, risk scores, and references are consistent throughout
3. **Check completeness**: All template sections should be populated or explicitly marked N/A
4. **Format cleanup**: Ensure markdown formatting is clean (tables aligned, headers consistent, no orphaned references)
---
## 3. Completion Summary
Report:
- Test files created
- Checklist output path
- Key risks or assumptions
- Next recommended workflow (e.g., implementation or `automate`)
---
## 4. Save Progress
**Save this step's accumulated work to `{outputFile}`.**
- **If `{outputFile}` does not exist** (first save), create it with YAML frontmatter:
```yaml
---
stepsCompleted: ['step-05-validate-and-complete']
lastStep: 'step-05-validate-and-complete'
lastSaved: '{date}'
---
```
Then write this step's output below the frontmatter.
- **If `{outputFile}` already exists**, update:
- Add `'step-05-validate-and-complete'` to `stepsCompleted` array (only if not already present)
- Set `lastStep: 'step-05-validate-and-complete'`
- Set `lastSaved: '{date}'`
- Append this step's output to the appropriate section.
## 🚨 SYSTEM SUCCESS/FAILURE METRICS:
### ✅ SUCCESS:
- Step completed in full with required outputs
### ❌ SYSTEM FAILURE:
- Skipped sequence steps or missing outputs
**Master Rule:** Skipping steps is FORBIDDEN.