Back to .md Directory
intent_id: INT-001 owner: agent-taskstate-team status: active last_reviewed_at: 2025-03-07 next_review_due: 2025-04-07
agent-taskstate Task Seed Template
メタデータ
task_id: YYYYMMDD-xx
repo: https://github.com/owner/agent-taskstate
base_branch: main
work_branch: feat/short-slug
priority: P1|P2|P3
langs: [python]
Objective
{{一文で目的}}
Scope
- In: {{対象(ディレクトリ/機能/CLI)を箇条書き}}
- Out: {{非対象(触らない領域)を箇条書き}}
Requirements
- Behavior:
- {{期待挙動1}}
- {{期待挙動2}}
- I/O Contract:
- Input: {{型/例}}
- Output: {{型/例}}
- Constraints:
- JSON契約
{ok, data, error}遵守 - typed_ref 形式使用
- Lint/Type/Test はゼロエラー
- JSON契約
- Acceptance Criteria:
- {{検収条件1}}
- {{検収条件2}}
Affected Paths
- {{glob例: docs/src/, tests/, src/agent_taskstate.py}}
Local Commands
# Python
ruff check . && black --check . && mypy src/ && pytest tests/ -v
# Fallback
python -m pytest tests/ -v --cov=src
Deliverables
- PR: タイトル/要約/影響/ロールバック
- 本文へ
Intent: INT-xxxと## EVALUATIONアンカーを明記
- 本文へ
- Artifacts: 変更パッチ、テスト、必要ならREADME/CHANGELOG差分
Plan
Steps
- 現状把握(対象ファイル列挙、既存テストとI/O確認)
- テストを先に記述
- 最小限の実装
- テスト実行→通過確認
- リファクタリング
- ドキュメント更新(必要なら)
Patch
Provide a unified diff. Include full paths. New files must be complete.
Tests
Outline
- Unit:
- {{case-1: 入力→出力の最小例}}
- {{case-2: エッジ/エラー例}}
- Integration:
- {{代表シナリオ1つ}}
Commands
Run gates
pytest tests/ -v
ruff check src/ tests/
black --check src/ tests/
mypy src/
Notes
Rationale
- {{設計判断を1~2行}}
Risks
- {{既知の制約/互換性リスク}}
Follow-ups
- {{後続タスクあれば}}
Template Usage
このテンプレートをコピーして TASK.<slug>-MM-DD-YYYY.md として保存:
# 例
TASK.task-create-test-03-07-2025.md
TASK.state-patch-conflict-03-07-2025.md
Related Documents
SPEC.md
Community AI Agent Skills Discovery Sources
**Research Date:** 2026-03-26
aiagentprompt
0
13
liqiongyuSPEC.md
GPU Selection Guide for Large Language Models (LLMs)
This guide helps you choose the right GPU for running Large Language Models, whether you're using them for inference, fine-tuning, or training.
aillm
0
12
mshojaei77SPEC.md
ReleaseKit - Technical Requirements Document
ReleaseKit provides **two interfaces** to the same underlying functionality:
aiagentllm
0
9
grokifySPEC.md
api_llm Specification
Provide direct, transparent HTTP API bindings for major LLM providers without abstraction layers or automatic behaviors.
aillmrag
0
8
iron-cage