A command for turning a broad software task into a structured, trackable project item. It checks the request, defines a goal and completion conditions, identifies major phases and risks, and can add the result to a task file after confirmation.

Use it to initialize a high-level task or project phase, provided the workspace has the required task file and the user has supplied a goal.

It helps clarify vague requests and record the work in a consistent project-planning format.

task-init — Command by Zenobia000

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/zenobia000/cursor-agentic-coding-template/task-init

Clone the repo

git clone --depth 1 https://github.com/Zenobia000/cursor-agentic-coding-template

Made for: Cursor.

Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,131 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00018	$0.01131
Opus 5	$0.00009	$0.00566
Sonnet 5	$0.00004	$0.00226
Haiku 4.5	$0.00002	$0.00113

Measured 2d ago against content hash 5cff098c61be, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-init scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

task-init — 100% identical, 0 lines differ

.cursor/commands/task-init.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

🚀 TASK INIT (v3)

為一次新的冒險啟航！將用戶的模糊想法，轉化為一個清晰、可追蹤的任務 Epic。

1. PLAN 🎯 (規劃)

Objective

將一個新的、高層次的用戶需求，轉化為一個結構化的 Epic 任務，並在用戶確認後，將其添加到 memory-bank/current/tasks.md 的頂部，作為後續 /plan 指令的輸入。

Guiding Rules

在執行此指令時，AI Agent 必須遵循以下規則：

主要規則: .cursor/rules/principles/global.mdc
規劃流程: .cursor/rules/process/planning-process.md
核心隔離: .cursor/rules/isolation_rules/main.mdc

Prerequisites Check

在初始化一個新任務之前，請確保：

✅ tasks.md 存在: 核心任務文件 memory-bank/current/tasks.md 已由 /van 指令創建。
✅ 需求已提供: 用戶已提出一個新的目標或需求。
❌ Failure Action: 如果 tasks.md 不存在，AI 應建議先運行 /van。

2. DO 実行 (執行)

Core Process

遵循一個從需求到 Epic 的結構化轉化流程：

Step 0: 健康檢查 (Health Check)

檢查需求清晰度: 分析用戶的請求。
批判性思考: 如果用戶的需求極度模糊 (e.g., "給我做個 AI")，AI 必須通過提問來引導用戶聚焦。

Step 1: 記憶體互動 - 讀取 (Memory Interaction - Read)

讀取用戶需求: 這是本指令的主要輸入。
(可選) 讀取專案簡報: 快速掃描 memory-bank/current/projectbrief.md，以確保新任務與專案大方向一致。

Step 2: 結構化分析 (Structured Analysis)

定義目標 (Goal): 用一句話總結任務的最終目標。
定義驗收條件 (DoD): 列出 3-5 個關鍵的、可衡量的「完成定義」。
初步分解 (WBS): 創建一個非常高層次的階段劃分。
風險評估 (Risks): 識別 1-3 個最顯而易見的風險。

Step 3: 記憶體互動 - 確認與寫入 (Memory Interaction - Confirm & Write)

向用戶確認 (Confirm): 將上述分析結果，以一個清晰的「任務提案」形式呈現給用戶，並請求批准。
追加寫入 tasks.md (Append): 在得到用戶肯定的答覆後，將這個結構化的 Epic 任務追加到 memory-bank/current/tasks.md 文件的頂部。注意：是追加 (append)，不是覆蓋 (overwrite)。

3. CHECK ✓ (檢查)

Verification Checklist

提案是否清晰: AI 生成的「任務提案」是否準確且簡潔地反映了用戶的需求？
用戶是否確認: AI 是否在寫入 tasks.md 之前，獲得了用戶的明確批准？
tasks.md 是否已更新: 檢查 memory-bank/current/tasks.md 文件，確認新的 Epic 任務是否已成功追加。

4. ACT 改善 (行動)

Finalization

報告完成: 在更新 tasks.md 後，向用戶報告。例如：「好的，我已經將『XXX』任務作為一個新的 Epic 添加到我們的計畫中。」

Next Steps

Epic 的創建是詳細規劃的開始。

👉 Primary Next Step: 執行 /plan 指令。/plan 指令會讀取剛剛創建的 Epic，並將其進一步分解為詳細的、可執行的子任務。
💡 Alternative: 如果識別出的風險非常高，可以先執行一個研究性質的任務 (Spike)，對不確定的技術點進行預研。

TASK INIT (v3) 已啟動。正在聆聽您的新想法，並準備將其轉化為行動藍圖...

Read the full file on GitHub · 68 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 68 lines · 18 tokens per session scan A 5cff098c61be

Subscribe to this mod's changes

task-init is a command published in the GitHub repository Zenobia000/cursor-agentic-coding-template (30 stars, last pushed 2mo ago), licensed MIT. It adds 18 tokens to every session and 1,131 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Planning

task-init