ll-qa-prototype-to-e2eplan

name	ll-qa-prototype-to-e2eplan
description	ADR-047 governed skill for deriving user journeys from a frozen prototype (or feat) and generating an E2E journey plan (e2e-journey-plan.md) with main journeys and minimum exception journeys.

LL QA Prototype to E2E Plan

This skill implements the ADR-047 E2E test chain entry point. It accepts a frozen prototype document (or a FEAT for API-derived mode) and produces an e2e-journey-plan.md that identifies user-visible journeys including at least one main journey and one exception journey. It is the first step in the E2E test chain: Prototype/FEAT → e2e-journey-plan → e2e-coverage-manifest → e2e-journey-specs.

Canonical Authority

• ADR: E:\ai\LEE-Lite-skill-first\ssot\adr\ADR-047-双链测试架构与防偷懒治理框架.MD
• Workflow template: E:\ai\LEE-Lite-skill-first\ssot\tests\templates\prototype-to-e2e-journey-plan.md
• Upstream handoff: ll-dev-feat-to-proto (frozen prototype) or ll-product-epic-to-feat (FEAT for API-derived mode)
• Downstream skill target: ll-qa-e2e-manifest-init

Runtime Boundary Baseline

• Interpret this workflow using E:\ai\LEE-Lite-skill-first\ssot\adr\ADR-038-运行时核心抽象边界与对象分层基线.MD.
• This capability is a governed Skill for Prototype -> E2E Journey Plan derivation.

Required Read Order

1. ll.contract.yaml
2. input/contract.yaml
3. output/contract.yaml
4. agents/executor.md
5. agents/supervisor.md
6. input/semantic-checklist.md
7. output/semantic-checklist.md

Execution Protocol

1. Accept one frozen prototype package from ll-dev-feat-to-proto, or a FEAT for API-derived mode.
2. Determine mode:
   • Prototype-Driven: prototype exists with flow map / journey map → extract journeys directly
   • API-Derived: no prototype → derive user-visible journeys from FEAT capabilities
3. Apply journey identification rules:

   触发条件	必须识别的旅程	优先级
   每个页面流	至少 1 条主旅程	P0
   每个表单页	至少 1 条校验失败异常旅程	P0
   每个有状态差异的页面	至少 1 条状态型旅程	P1
   每个有回访入口的闭环	至少 1 条 revisit 旅程	P1
   每个主旅程	至少 1 条关键分支旅程	P1
   每个网络请求点	至少 1 条网络失败异常旅程	P1

4. For each journey, define:
   • journey_id (format: JOURNEY-{TYPE}-{SEQ}, e.g., JOURNEY-MAIN-001)
   • journey_type (main, exception, branch, retry, state)
   • entry_point (page/URL where journey begins)
   • user_steps (ordered list of user actions)
   • expected_ui_states (UI states at key checkpoints)
   • expected_network_events (API calls expected during journey)
   • expected_persistence (data that must be saved/changed)
   • priority (P0/P1)
5. Validate minimum journey count: ≥1 main journey (P0) + ≥1 exception journey.
6. Generate e2e-journey-plan.md with:
   • Plan metadata (prototype_id or feature_id, derivation_mode, created_at, source, anchor_type=prototype)
   • Journey definitions table
   • Journey detail sections for each journey
   • Minimum journey count validation result
7. Run executor agent to draft the plan, then supervisor agent to validate journey completeness.
8. Emit the plan to ssot/tests/e2e/{prototype_id}/e2e-journey-plan.md.

Workflow Boundary

• Input: one frozen prototype package OR one frozen FEAT (API-derived mode)
• Output: one e2e-journey-plan.md at ssot/tests/e2e/{prototype_id}/
• Out of scope: generating manifests (downstream skill), generating journey specs (downstream skill), running actual E2E tests

Non-Negotiable Rules

• Do not accept non-frozen prototype or FEAT documents.
• Do not generate a plan with zero exception journeys — minimum 1 required.
• Do not generate a plan with zero main journeys — minimum 1 required.
• Do not skip network failure exception journeys for journeys that make API calls.
• Do not let API-derived mode skip the derivation_mode: api-derived annotation.
• Every journey must have an entry_point and at least 2 user_steps.
• Do not modify prototype or FEAT content during journey extraction.