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.
curl -O https://raw.githubusercontent.com/receptron/mulmoclaude/main/.claude/skills/make-e2e-live/SKILL.mdgit clone --depth 1 https://github.com/receptron/mulmoclaudeWrote 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.
[](https://agentmods.dev/skills/receptron/mulmoclaude/make-e2e-live)<a href="https://agentmods.dev/skills/receptron/mulmoclaude/make-e2e-live"><img src="https://agentmods.dev/badge/skills/receptron/mulmoclaude/make-e2e-live/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.
<a href="https://agentmods.dev/skills/receptron/mulmoclaude/make-e2e-live"><img src="https://agentmods.dev/badge/skills/receptron/mulmoclaude/make-e2e-live.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00103 | $0.05804 |
| Opus 5.5 | $0.00041 | $0.02322 |
| Sonnet 5.5 | $0.00021 | $0.01161 |
| Haiku 4.5 | $0.00010 | $0.00580 |
Grade A, and why
make-e2e-live 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.
この skill の位置づけ
| skill | 役割 |
|---|---|
/e2e-live |
既存スイートを 回す（QA / 回帰検出） |
/e2e-live-<category> |
既存カテゴリだけ回す |
/make-e2e-live（このskill） |
既存スイートを 育てる（未実装シナリオ追加 / 修正 / config 改善） |
「実 LLM e2e の追加実装をしたい」 と思ったらこの skill を起点にする。 巨大 PR を避けるため 1 PR = 1〜3 シナリオ or 1 config 改善 に絞ること。
Phase 1: 状況把握
最初に以下をユーザーに見せる前に自分で読む:
plans/feat-e2e-live.mdの以下セクション:- 「実装ステータス」 表（L-01〜L-30 のうち ✅ / 未実装の現状）
- 「直近 main の動向 (#950〜#1000) と本テスト計画への反映」 の 「要対応」 項目
- 「未確定事項 / TODO」 のチェックリスト
plans/survey-e2e-live-test-step-and-nonce.mdの以下:- 「調査タスク」 1〜6 の進捗（plan ファイル上のチェック有無）
- 「各 spec の見立て」 表 —
test.step化 /testInfo.titlenonce 化の横展開候補（PR #1347 で L-15b にのみ適用済、 他 spec への横展開が残 TODO）
- main が最新かを確認 → 必要な場合のみ最新化:
# まず remote の最新を取り込む（fetch のみ、 作業ツリーは触らない） git fetch origin main # ローカル main が origin/main に対して何コミット遅れているか git rev-list --count main..origin/main- 出力が
0→ 既に最新。 pull は不要、 そのまま次のステップへ - 出力が
1以上 → 遅れているので最新化:git checkout main && git pull --ff-only
- 出力が
- 前回 e2e-live PR merge 以降の main の動きを確認:
spec の前提を変える PR があれば優先順位を上げる。 過去例:git log main --oneline --since="<前回 merge 日>"- #969 / #972 image-path-routing → L-01 assertion 更新
- #982 filePath-only /
/artifacts/htmlmount → L-01 prompt convention 更新 - #974 onerror self-repair → L-01
naturalWidth > 0の検出力低下、 緩和策が TODO 残 - #991 Safari preview iframe CSP → webkit project 追加が TODO 残
- ユーザーから chat で追加要望が来ていれば、 上記 4 つに統合する。
Phase 2: 着手項目の自動選定
Phase 1 の結果から、 以下ルールで 1 PR 分の着手項目を自動選定する。 確認待ちで止まらない（auto mode 親和、 ユーザーは Phase 3〜4 進行中いつでも 「やっぱり L-04 で」 と redirect 可）。
PR の規模（1 シナリオ vs 数シナリオ）はユーザー判断。 デフォルトは 1 シナリオ単位、 ユーザーが「カテゴリ別 1 つずつ採取」 や「5-6 個一気に」 と指示したらそれに従う。
選定優先度（上から順に試す）
- C. ユーザー追加要望 — skill 起動プロンプトに項目指定（「L-03 で」 「webkit project 追加」 「カテゴリ別 1 つずつ」 等）があれば最優先で採用
- B. config / 基盤改善 — main 動向の 「要対応」 に未消化があれば 1 つ（webkit project / self-repair 緩和 等）、 もしくは
plans/survey-e2e-live-test-step-and-nonce.mdの test infra 横展開（test.step/ nonce 化）から 1 spec。 規模が小さく副作用も spec 不変で reversible なので先に消化 - A. 未実装シナリオ —
plans/feat-e2e-live.mdの 「実装ステータス」 で未実装のうち、 重要度 S → A → B 順、 同レベル内では番号若い順で 1 つ
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.
- 1mo ago First seen · 255 lines · 103 tokens per session scan A 4a65e120d873
make-e2e-live is a skill published in the GitHub repository receptron/mulmoclaude (386 stars, last pushed 2d ago), licensed MIT. It adds 103 tokens to every session and 5,804 once invoked, about $0.0004 per session on Opus 5.5. 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.
Other skills, from other repositories
develop-web-game
Use when Codex is building or iterating on a web game (HTML/JS) and needs a reliable development + testing loop: implement small changes, run a Playwright-based test script with short input bursts and intentional pauses, inspect screenshots/text, and review console errors with rendergametotext.
startup-qa-bot-getting-started
A guided first conversation for a product-checking assistant. It helps define a safe test environment and choose an initial check, such as following a sign-up flow or comparing two versions.
startup-qa-bot-ops
A working procedure for checking a product in a clearly separated test environment. It maintains a page map, records observations, compares known page lists, and reports reproducible problems.
proto-bot-evidence
A small local checker that compares acceptance requirements with evidence for required interface themes or modes. It checks supplied data but does not render pages or prove that a person inspected them in a browser or on a device.
product-test
Use this skill to define the observable behavior that must be true for a story or feature to be considered done. It turns product context into concise, testable Given/When/Then scenarios that engineers and QA can verify without guessing intent.
startup-qa-bot-evidence
An offline comparison tool for two supplied page lists from the same test environment. It reports differences and status changes without opening a browser.