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.
npx skills add masa-san-jp/Agent-Aiko --skill live-previewgit clone --depth 1 https://github.com/masa-san-jp/Agent-AikoWrote 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/masa-san-jp/agent-aiko/live-preview)<a href="https://agentmods.dev/skills/masa-san-jp/agent-aiko/live-preview"><img src="https://agentmods.dev/badge/skills/masa-san-jp/agent-aiko/live-preview/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/masa-san-jp/agent-aiko/live-preview"><img src="https://agentmods.dev/badge/skills/masa-san-jp/agent-aiko/live-preview.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00171 | $0.01159 |
| Opus 5 | $0.00086 | $0.00580 |
| Sonnet 5 | $0.00034 | $0.00232 |
| Haiku 4.5 | $0.00017 | $0.00116 |
Grade C, and why
live-preview scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
## 有効化(allowlist のみ・sudo 不要) Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
tailscale serve を使わず tailscale IP に直接 bind する方式なので **sudo/operator/HTTPS 証明書は不要**。唯一必要なのは、本スクリプトをエージェントの権限 allowlist に入れること(ネットワーク配信を伴うため)。設定ファイル(例: `{{ORG_REPO_PATH}}/.claude/settings.json` の `permissions.allow`)に、自分の配置 How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/live-preview
エージェントが作った HTML/静的成果物を、tailnet 内だけで見れる「ライブ URL」として出す共通スキル。汎用スキル tailnet-expose の HTML 特化ラッパー。
何のためか
ユーザーが別端末(スマホ等)から成果物を見る場合、file:// パスやこのマシンの localhost は届かない。作った HTML を本人の端末から URL で見れて初めて「見て決める」が回る。Tailscale の tailnet 内に出せば、同じ tailnet の端末からだけ見れて、外部には出ない。これを全エージェントが 1 コマンドで使える共通スキルにする。
不可侵ルール
- tailnet 内のみ。
tailscale funnel(インターネット公開)は絶対に使わない。 成果物には未公開情報が含まれうる。露出範囲はユーザー本人の tailnet 端末に限定する。 - bind 先は tailscale IP に限定する(
0.0.0.0/LAN IP に bind しない)。tailnet 外には出さない。 - 使い終わったら
stopする。 配信を出しっぱなしにしない。次の成果物を出す前に前のを片付ける。 - 個人情報・機微を載せた成果物を渡すときも、宛先はユーザー本人の端末のみ(tailnet)であることを前提にする。第三者の実名等は成果物自体に入れない。
使い方
同梱スクリプト {{ORG_REPO_PATH}}/Agent-team/tools/live-preview/live-preview.sh を呼ぶ。
SK={{ORG_REPO_PATH}}/Agent-team/tools/live-preview/live-preview.sh
bash "$SK" serve <path-to-html-or-dir> # → ライブ URL(http://<host>.<tailnet>.ts.net:<port>/...) を返す
bash "$SK" status # 現在の配信状態
bash "$SK" stop # 配信停止(必ず最後に)
返ってきた URL をそのままユーザーの端末へ送る。ディレクトリを渡せばその中の index.html が入口。ファイルを渡せば …/<file> が直接開く。
有効化(allowlist のみ・sudo 不要)
tailscale serve を使わず tailscale IP に直接 bind する方式なので sudo/operator/HTTPS 証明書は不要。唯一必要なのは、本スクリプトをエージェントの権限 allowlist に入れること(ネットワーク配信を伴うため)。設定ファイル(例: {{ORG_REPO_PATH}}/.claude/settings.json の permissions.allow)に、自分の配置に合わせた絶対パスで登録する:
"Bash(bash {{ORG_REPO_PATH}}/Agent-team/tools/live-preview/live-preview.sh:*)"
これで全エージェントが serve 一発でライブ URL を出せる。
URL の形
http://<host>.<tailnet>.ts.net:<port>/。MagicDNS が引ければホスト名、引けなければ tailscale IP を使う。HTTP(HTTPS でない)だが tailnet 内に閉じているため preview 用途では十分。
制約
- 実体は汎用スキル
tailnet-expose。本スクリプトはその HTML 特化ラッパーで、内部でtailnet-exposeのserveを呼ぶ。 - 成果物ごとに配信を切り替えるため、前の配信を
stopしてから次をserveする。
関連
- 汎用版:
{{ORG_REPO_PATH}}/Agent-team/reference/skills/tailnet-expose/SKILL.md(/tailnet-expose) - 実装本体:
{{ORG_REPO_PATH}}/Agent-team/tools/live-preview/live-preview.sh
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.
- 9d ago First seen · 57 lines · 171 tokens per session scan C 2ebe6a98a2d6
live-preview is a skill published in the GitHub repository masa-san-jp/Agent-Aiko (3 stars, last pushed 1mo ago), licensed MIT. It adds 171 tokens to every session and 1,159 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it C with 2 findings (asks for root, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
harness-fe
Debug, inspect, and drive any frontend app that has the Harness-FE Vite/Webpack plugin installed. Use this when the user reports a UI bug, asks "why is this happening on the page", wants to inspect runtime state, drive an automated browser test, or needs to correlate browser behavior with source files (especially in…
cadis-ramaclaw-ui
Use when implementing or reviewing CADIS HUD, RamaClaw UI adaptation, config window, agent rename, theme picker, voice/model settings, orbital HUD, approval cards, screenshot parity, or desktop window behavior.
frontend-design
Design distinctive, polished UI that avoids generic 'AI slop'. Use when making visual design decisions: typography, color, spacing, hierarchy, motion, and component composition.
responsive-check
Basic responsive design checks.
accessibility
WCAG 2.1 AA compliance audit.
visual-ralph
Visual Ralph orchestration for frontend UI from generated references, static references, or live URL targets, using $ralph with built-in visual verdict and pixel-diff evidence until the implementation matches and leaves a reproducible design system.