mulmoclaude: Skill for Claude Code

.claude/skills/e2e-live-docker/SKILL.md

e2e-live-docker is a skill for Claude Code from receptron/mulmoclaude. It costs 85 tokens per session (1,001 once invoked), scanned A, original, MIT.

A live end-to-end test suite for checking a coding agent running inside a Docker sandbox. It uses the real Claude API and checks how the sandbox works with authentication, saved sessions, and command-line tools.

In plain words
What is it for?
Run the Docker-based tests with `yarn test:e2e:live:docker`. Use it to check X tools, session recovery after a page reload, and GitHub CLI authentication inside the container.
Why use it?
It helps catch failures that ordinary tests may miss when the agent, Docker container, host credentials, and web app must work together. Tests are skipped when the required local services, credentials, or settings are missing.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is receptron/mulmoclaude's own configuration. It tells Claude Code how to work on mulmoclaude 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 mulmoclaude configures →

Reuse

Borrowing it

Nothing to install: this file belongs to receptron/mulmoclaude. 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/receptron/mulmoclaude/main/.claude/skills/e2e-live-docker/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/receptron/mulmoclaude

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 e2e-live-docker

README.md
[![agentmods](https://agentmods.dev/badge/skills/receptron/mulmoclaude/e2e-live-docker/github.svg)](https://agentmods.dev/skills/receptron/mulmoclaude/e2e-live-docker)
Your own site
<a href="https://agentmods.dev/skills/receptron/mulmoclaude/e2e-live-docker"><img src="https://agentmods.dev/badge/skills/receptron/mulmoclaude/e2e-live-docker/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 e2e-live-docker

Your own site · 80×15
<a href="https://agentmods.dev/skills/receptron/mulmoclaude/e2e-live-docker"><img src="https://agentmods.dev/badge/skills/receptron/mulmoclaude/e2e-live-docker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 85 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,001 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium MCP Rug Pull · line 37
    npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
    Fix: Pin the version: npx @scope/[email protected]
How audits are shown
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 $0.00085 $0.01001
Opus 5 $0.00043 $0.00500
Sonnet 5 $0.00017 $0.00200
Haiku 4.5 $0.00009 $0.00100

Measured 10d ago against content hash 0ea92ca1f034, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

e2e-live-docker 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 10d 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.

.claude/skills/e2e-live-docker/SKILL.md · 46 lines

What it actually says

前提

  • yarn dev を別ターミナルで起動済み(http://localhost:5173 が応答する)
  • DISABLE_SANDBOX を unset で起動していること。/api/sandbox の payload は enabled 時 { sshAgent, mounts }、 disabled 時 空オブジェクト {} を返す(server/api/sandboxStatus.ts 参照)。spec 側の getSandboxStatus(page) helper はこの {}null に正規化するので、 sandbox off だと spec の gate が null === null で hit して全 test が test.skip で抜ける
  • Docker daemon が起動しており、 mulmoclaude-sandbox image が build 済(yarn dev 初回起動時に自動 build)
  • Claude 認証済み(claude login 済み or ANTHROPIC_API_KEY 設定済み)
  • 各シナリオ固有の前提(unmet なら spec が test.skip で抜ける、 ログにスキップ理由が出る):
    • L-23: X_BEARER_TOKEN が host の env に設定されていること
    • L-28: SANDBOX_MOUNT_CONFIGS=gh または SANDBOX_SSH_AGENT_FORWARD=1 のいずれかが設定されていること

実行

yarn test:e2e:live:docker

デバッグ時

HEADED=1 yarn test:e2e:live:docker

カバーするシナリオ

  • L-23: X MCP tools (readXPost / searchX) が Docker サンドボックス on 状態で host の X_BEARER_TOKEN を見て enabled になる(B-01 回帰)
  • L-26: サンドボックス on 状態で開始したセッションが reload で「No conversation found」 を出さず履歴復元される(B-04 回帰、 in-container workspace path 整合)
  • L-28: agent が container 内で gh auth status を Bash 実行し、 host の gh credential が container に届いていることを確認する(B-06 回帰)

結果の確認

  • 詳細: playwright-report-live/docker/index.html(このカテゴリ専用サブディレクトリに出力されるので、 親 /e2e-live の総合レポートは上書きされない)
  • 動画リプレイ: npx playwright show-trace test-results-live/docker/<spec>/trace.zip

fake-echo / CI matrix について

このカテゴリは fake-echo backend (MULMOCLAUDE_FAKE_AGENT=1) で再現できない(実 Docker サンドボックス起動が必要)ので、 .github/workflows/e2e_live_no_llm.yaml の matrix には 意図的に登録していない。 同等の理由で各 test も E2E_LIVE_NO_LLM=1 が立っていると skip する。 routine な CI 検証ではなく 開発者が手元で Docker on モードを切り替えて回す ことを想定したスイート。

Docker on/off の検証も必要な場合

このカテゴリは Docker on 必須で、 off では全 test が test.skip するだけ。 「両モード巡回」 を取りたい時は 1) 現在モード (/api/sandbox の payload で確認) で yarn test:e2e:live:docker を回す、 2) もう一方のモードに dev を再起動して yarn test:e2e:live を回して他カテゴリの両モード健全性を見る、 の併用が想定。

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. 10d ago First seen · 46 lines · 85 tokens per session scan A 0ea92ca1f034

Subscribe to this mod's changes

e2e-live-docker is a skill published in the GitHub repository receptron/mulmoclaude (346 stars, last pushed yesterday), licensed MIT. It adds 85 tokens to every session and 1,001 once invoked, about $0.0004 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.