M3E: Skill for Claude Code

.claude/skills/canvas-protocol/SKILL.md

canvas-protocol is a skill for Claude Code from akaghef/M3E. It costs 266 tokens per session (2,567 once invoked), scanned C, original, MIT.

A protocol for using a shared online whiteboard as a coordination space between developers and coding agents. It defines where decisions, questions, task status, ideas, and bugs are stored, while allowing implementation to continue if the board is unavailable.

In plain words
What is it for?
Use it to read pending decisions before starting work, record progress, reorder tasks, and share review questions or bug reports with a team.
Why use it?
It prevents agents from missing human decisions or duplicating coordination work. It also keeps the project from depending on the whiteboard for the actual code changes.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents; names the AskUserQuestion tool.

This is akaghef/M3E's own configuration. It tells Claude Code how to work on M3E itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything M3E configures →

View source ↗ akaghef/M3E
Reuse

Borrowing it

Nothing to install: this file belongs to akaghef/M3E. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/akaghef/M3E/main/.claude/skills/canvas-protocol/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/akaghef/M3E

Made for: Claude Code.

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 canvas-protocol

README.md
[![agentmods](https://agentmods.dev/badge/skills/akaghef/m3e/canvas-protocol/github.svg)](https://agentmods.dev/skills/akaghef/m3e/canvas-protocol)
Your own site
<a href="https://agentmods.dev/skills/akaghef/m3e/canvas-protocol"><img src="https://agentmods.dev/badge/skills/akaghef/m3e/canvas-protocol/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for canvas-protocol

Your own site · 80×15
<a href="https://agentmods.dev/skills/akaghef/m3e/canvas-protocol"><img src="https://agentmods.dev/badge/skills/akaghef/m3e/canvas-protocol.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 266 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,567 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00266 $0.02567
Opus 5.5 $0.00106 $0.01027
Sonnet 5.5 $0.00053 $0.00513
Haiku 4.5 $0.00027 $0.00257

Measured 5d ago against content hash 3f51df285df3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade C, and why

canvas-protocol scanned grade C with 2 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 5d 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

curl -s http://localhost:4173/api/maps | node -e "..."

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s http://localhost:4173/api/maps | node -e "..."
.claude/skills/canvas-protocol/SKILL.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

canvas-protocol — Agent ↔ Canvas Interaction Rules

M3E マップを共有 canvas として agent と人間で使う。canvas は 任意の同期ポイントであり、 agent の主作業 (実装・テスト・PR) は canvas 不在でも完結できる設計を保つ。

1. Target server

  • default: beta (port 4173)
  • map discovery: GET /api/maps で label="開発" の mapId を取得
  • final (38482) は本番確認時のみ使う (m3e-map skill 参照)

2. Canvas Layout (固定パス)

M:(開発)> SYSTEM > DEV
├── strategy/                       # 開発戦略・タスクボード
│   └── HOME Re-implementation/     # プロジェクト単位のサブツリー
│       ├── Visual tasks/           # ロール別タスク
│       ├── Data tasks/
│       └── Team coordination/
├── reviews/                        # 人間の判断待ちキュー
│   └── {Project}/Q1, Q2, ...       # 各 Q の子ノードが選択肢
│       └── (option) attributes.selected="yes" で確定
├── Agent Status/                   # 各 worker の現在状態
├── tasks/ doing / ready / done     # devM3E の Phase 5 同期先
├── scratch/                        # アイデア・バグ報告・後回し
└── Bugs/                           # 既知バグ

3. Read Protocol (agent → canvas)

着手前に必ず実施:

# reviews/{Project} を読み、status="open" の Q に未回答が無いか確認
curl -s http://localhost:4173/api/maps | node -e "..."

判断:

  • 該当 Q が attributes.selected="yes" 付き選択肢を持つ → その決定で進める
  • status="open" で未選択 → 待機 or 自己解釈せず人間に確認
  • status="resolved" だが selected が無い → 人間に再確認 (canvas が壊れている可能性)

重要: canvas に到達できない / fetch 失敗の時、agent は作業を止めない。 ローカルで実装ドラフトを進め、PR description に「canvas 未参照、要確認」と明記する。

4. Write Protocol (agent → canvas)

4.1 タスク開始

// strategy/{Project}/{Role} tasks/{taskNode}.attributes.status = "in_progress"
node.attributes = {...node.attributes, status: "in_progress", startedAt: "2026-04-14"};

4.2 タスク完了

node.attributes = {...node.attributes, status: "done", completedAt: "2026-04-14"};

4.3 新規 Q 起票 — Ambiguity Pooling (MUST)

原則: 曖昧点に当たったら block しない / silently 決めない。プールして暫定値で進む。akaghef が batch-review する。

reviews/{Project}/Qn (text="Qn: <一文の質問>",
  attributes={status:"open", raisedBy:"<role>", raisedAt:"YYYY-MM-DD"})
└── 選択肢 1 (details: Why / Trade-off)
└── 選択肢 2 (details: Why / Trade-off)
└── 選択肢 3 (attributes.tentative="yes" ← agent の暫定採用案)

Read the full file on GitHub · 169 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. 5d ago First seen · 169 lines · 266 tokens per session scan C 3f51df285df3

Subscribe to this mod's changes

canvas-protocol is a skill published in the GitHub repository akaghef/M3E (11 stars, last pushed 5d ago), licensed MIT. It adds 266 tokens to every session and 2,567 once invoked, about $0.0011 per session on Opus 5.5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-02.