ワンクリックで
long-task-finalize
Use after ST Go verdict — generate usage examples and finalize release documentation via SubAgent
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Use after ST Go verdict — generate usage examples and finalize release documentation via SubAgent
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when design doc exists but no ATS doc and no feature-list.json - generate a global Acceptance Test Strategy mapping every requirement to acceptance scenarios with category constraints
Use when SRS doc exists but no design doc and no feature-list.json - take the approved SRS as input and produce an architecture/design document focused on HOW to build it
Use when ATS doc exists (or auto-skipped) but feature-list.json not yet created - scaffold project artifacts and populate features from Design §10.2
Use when no SRS doc and no design doc and no feature-list.json exist - elicit requirements through structured questioning and produce a high-quality SRS document aligned with ISO/IEC/IEEE 29148
Use when starting any session in a long-task project - routes to the correct phase skill based on project state
Use when increment-request.json exists - collect incremental requirements, perform impact analysis, update design, and decompose new features
| name | long-task-finalize |
| description | Use after ST Go verdict — generate usage examples and finalize release documentation via SubAgent |
Generate scenario-based usage examples and finalize release documentation after System Testing passes with a Go/Conditional-Go verdict.
Announce at start: "I'm using the long-task-finalize skill. ST passed — generating examples and finalizing documentation."
Idempotent: Safe to re-invoke after ST defect-fix loops. Overwrites examples/ content cleanly on each run.
You MUST create a TodoWrite task for each step and complete them in order:
feature-list.json — all passing non-deprecated features, tech_stack, quality_gatesdocs/plans/*-srs.md) — requirement descriptions, user personasdocs/plans/*-design.md) — architecture, public API surfacedocs/plans/*-ucd.md) — only if UI features existtask-progress.md — session history for ST summary entryRELEASE_NOTES.md — current state for version entryDispatch the example-generator SubAgent to produce scenario-based usage examples.
Construct SubAgent prompt:
You are an Example Generator SubAgent.
## Your Task
1. Read the agent definition: Read <skills_root>/../agents/example-generator.md
2. Follow the process to generate scenario-based usage examples
3. Return your result using the Structured Return Contract
## Input Parameters
- feature-list.json: <path>
- SRS: <srs_path>
- Design: <design_path>
- UCD: <ucd_path> (or "none")
- tech_stack: <tech_stack_json>
- Working directory: <project_root>
Dispatch:
Agent(
description = "Generate usage examples for all features",
prompt = [constructed prompt]
)
Parse return contract:
Record in task-progress.md:
- Examples: <verdict> — N scenarios, N examples generated, N features covered
Add ST completion and version entry (moved from ST Persist):
[Unreleased] or create versioned section if appropriateAdd ST session summary entry (moved from ST Persist):
git add examples/ RELEASE_NOTES.md task-progress.md
git commit -m "docs: finalize release — examples, release notes, progress update"
python scripts/validate_features.py feature-list.json
Output completion summary:
Finalize — DONE
Examples: N scenarios generated (N features covered, N skipped) RELEASE_NOTES.md: Updated with ST completion task-progress.md: Updated with ST session summary
Called by: long-task-st (Step 13, after Go/Conditional-Go verdict)
Reads: feature-list.json, docs/plans/*-srs.md, docs/plans/*-design.md, docs/plans/*-ucd.md (if UI), task-progress.md, RELEASE_NOTES.md, implementation code
Produces: examples/ (usage examples + README.md), updated RELEASE_NOTES.md, updated task-progress.md
Agent: agents/example-generator.md