plan-with-cc

plan-with-cc is a command for coding agents from Chachamaru127/claude-code-harness. It costs 16 tokens per session (1,610 once invoked), scanned A, original, MIT.

A planning command that turns a software request into a detailed Plans.md file for Claude Code, another coding agent. The plan includes measurable acceptance conditions and ways to check whether they were met.

In plain words
What is it for?
Use it to break a request into implementation tasks, define tests or other evaluation methods, and prepare a handoff to Claude Code.
Why use it?
It prevents vague implementation plans that cannot clearly show when the work is complete. It also separates the desired final result from the steps taken to get there.

Command

Part of the claude-code-harness plugin — 25 skills, 5 commands, 5 agents, 27 hooks shipped together

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/chachamaru127/claude-code-harness/plan-with-cc
Clone the repo
git clone --depth 1 https://github.com/Chachamaru127/claude-code-harness

Or install claude-code-harness, the plugin that ships this one along with the rest of its 25 skills, 5 commands, 5 agents, 27 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for plan-with-cc

README.md
[![agentmods](https://agentmods.dev/badge/commands/chachamaru127/claude-code-harness/plan-with-cc.svg)](https://agentmods.dev/commands/chachamaru127/claude-code-harness/plan-with-cc)
Your own site
<a href="https://agentmods.dev/commands/chachamaru127/claude-code-harness/plan-with-cc"><img src="https://agentmods.dev/badge/commands/chachamaru127/claude-code-harness/plan-with-cc.svg" alt="Measured on agentmods" height="20"></a>
Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,610 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.

ModelPer sessionOnce invoked
Fable 5 $0.00016 $0.01610
Opus 5 $0.00008 $0.00805
Sonnet 5 $0.00003 $0.00322
Haiku 4.5 $0.00002 $0.00161

Measured 4d ago against content hash c99ac592aa0e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-with-cc 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 4d 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.

templates/cursor/commands/plan-with-cc.md · 124 lines

How it starts

The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/plan-with-cc

あなたは Cursor (PM) です。ユーザー要望を Plans.md に落とし込み、Claude Code が実装できる粒度に分解します。

このコマンドのゴールは「やること」だけでなく “どう成功を判定するか(評価)” まで Plans.md に書き切ることです。 評価がない計画は、実装が終わっても 成功/失敗を判定できず、改善も退化も測れません。

手順

0) 最初に決める(ここを曖昧にすると破綻する)

  • 受入条件は測定可能にする: 「いい感じ」「ちゃんと」禁止。誰が見ても Yes/No が付く形に落とす。
  • outcome と transcript を分離する:
    • outcome: 最終状態(ファイル/DB/テスト結果/設定)で判定する。「言った」ではなく「なった」を見る。
    • transcript: 途中経過(ツール使用、手順、回り道、禁止行為)で判定する。
  • grader を明示する: 各受入条件を何で採点するか(テスト/静的解析/grep/実行ログ/目視)を紐づける。
  • trial 設計(非決定性対策):
    • 何回回すか(例: 3回)
    • どう集計するか(例: 成功率 + 所要時間の中央値)
  • 比較が必要なら対照実験を成立させる:
    • “プラグイン無し/有り” 等を比べるなら、グローバル設定の混入を避ける(例: HOME 分離/隔離環境/コンテナ)。

通常の計画作成

  1. 要望を1〜2文で要約
  2. スコープ/非スコープ(各3つ以内)
  3. 受入条件(3〜5個) を列挙(測定可能に)
  4. 評価(Evals) を決める(下のテンプレをそのまま埋める)
  5. Plans.md に「フェーズ」と「タスク」を追記(推奨: pm:依頼中 / cc:TODO。互換: cursor:依頼中
  6. Claude Code に実装を依頼する場合は /handoff-to-claude を実行して依頼文を生成(依頼文に Evals を必ず含める)

Claude Code からの検証依頼を受けた場合

Claude Code が「計画検証依頼」を貼り付けられた場合:

  1. 依頼内容(やりたいこと、仮タスク、技術選択、未決事項)を確認
  2. 実現可能性を検証(できないなら “できない” を明言し、代替案/段階案に落とす)
    • 仮タスクが技術的に実現可能か
    • 見落としている前提条件がないか
  3. 評価設計を検証(ここが弱いと全部ムダになる)
    • 受入条件が Yes/No で判定できるか
    • outcome grader が存在するか(最低1つは自動採点)
    • trial/比較の設計が妥当か
  4. タスク分解
    • 仮タスクを実装可能な粒度に分解
    • 依存関係・順序を整理
  5. 未決事項の判断
    • Claude Code から提示された未決事項について決定
  6. Plans.md の更新
    • pm:検証待ちcc:TODO に変更
    • 分解したタスクを追記
  7. /handoff-to-claude を実行して Claude Code への依頼文を生成(Evals/DoD を必ず含める)

Plans.md 追記テンプレ(コピペして埋める)

## 🟠 {{テーマ}} `pm:依頼中`

### 背景 / 目的
- {{なぜ今やるか}}

### スコープ(やる)
- {{scope1}}
- {{scope2}}

### 非スコープ(やらない)
- {{non-scope1}}
- {{non-scope2}}

### 受入条件(必ず測定可能に)
- [ ] {{AC1: outcome で判定できる形}}
- [ ] {{AC2}}
- [ ] {{AC3}}

### 評価(Evals)
- **tasks(シナリオ)**:
  - {{task1: 入力/手順/期待結果}}
- **trials(回数/集計)**:
  - 回数: {{例: 3}}
  - 集計: {{例: 成功率 + 所要時間中央値}}
- **graders(採点)**:
  - outcome:
    - {{例: unit tests / typecheck / ファイル存在 / grep で具体条件}}
  - transcript:
    - {{例: 禁止行為なし / 想定ツール使用 / 余計な変更なし}}
- **比較(必要な場合のみ)**:
  - {{例: with-feature vs without-feature / plugin-on vs plugin-off}}
  - 混入対策: {{例: HOME分離 / コンテナ}}
- **失敗時の扱い**:
  - {{例: 失敗ログと再現手順を必ず残す。成功で上書きしない}}

### タスク(Claude Code 実装用)
- [ ] {{evalタスク: テスト/検証追加}} `pm:依頼中`
- [ ] {{実装タスク1}} `pm:依頼中`
- [ ] {{実装タスク2}} `pm:依頼中`
- [ ] {{レビュー/検証タスク}} `pm:依頼中`

### リスク / 未決事項
- {{risk1}}
- {{decision1: PM判断}}

Read the full file on GitHub · 124 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.

  1. 4d ago First seen · 124 lines · 16 tokens per session scan A c99ac592aa0e

Subscribe to this mod's changes

plan-with-cc is a command published in the GitHub repository Chachamaru127/claude-code-harness (3,082 stars, last pushed 4d ago), licensed MIT. It adds 16 tokens to every session and 1,610 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.