val-town
Write and deploy server-side TypeScript functions instantly with Val Town. Use when someone asks to "deploy a function quickly", "serverless TypeScript", "quick API endpoint", "webhook handler", "cron job in the cloud", "Val Town", "instant API without infrastructure", or "deploy a script without a server". Covers HTTP vals, cron vals, email vals, SQLite storage, and the Val Town API.
下記のコマンドをコピーしてターミナル(Mac/Linux)または PowerShell(Windows)に貼り付けてください。 ダウンロード → 解凍 → 配置まで全自動。
mkdir -p ~/.claude/skills && cd ~/.claude/skills && curl -L -o val-town.zip https://jpskill.com/download/15529.zip && unzip -o val-town.zip && rm val-town.zip
$d = "$env:USERPROFILE\.claude\skills"; ni -Force -ItemType Directory $d | Out-Null; iwr https://jpskill.com/download/15529.zip -OutFile "$d\val-town.zip"; Expand-Archive "$d\val-town.zip" -DestinationPath $d -Force; ri "$d\val-town.zip"
完了後、Claude Code を再起動 → 普通に「動画プロンプト作って」のように話しかけるだけで自動発動します。
💾 手動でダウンロードしたい(コマンドが難しい人向け)
- 1. 下の青いボタンを押して
val-town.zipをダウンロード - 2. ZIPファイルをダブルクリックで解凍 →
val-townフォルダができる - 3. そのフォルダを
C:\Users\あなたの名前\.claude\skills\(Win)または~/.claude/skills/(Mac)へ移動 - 4. Claude Code を再起動
⚠️ ダウンロード・利用は自己責任でお願いします。当サイトは内容・動作・安全性について責任を負いません。
🎯 このSkillでできること
下記の説明文を読むと、このSkillがあなたに何をしてくれるかが分かります。Claudeにこの分野の依頼をすると、自動で発動します。
📦 インストール方法 (3ステップ)
- 1. 上の「ダウンロード」ボタンを押して .skill ファイルを取得
- 2. ファイル名の拡張子を .skill から .zip に変えて展開(macは自動展開可)
- 3. 展開してできたフォルダを、ホームフォルダの
.claude/skills/に置く- · macOS / Linux:
~/.claude/skills/ - · Windows:
%USERPROFILE%\.claude\skills\
- · macOS / Linux:
Claude Code を再起動すれば完了。「このSkillを使って…」と話しかけなくても、関連する依頼で自動的に呼び出されます。
詳しい使い方ガイドを見る →- 最終更新
- 2026-05-18
- 取得日時
- 2026-05-18
- 同梱ファイル
- 1
📖 Claude が読む原文 SKILL.md(中身を展開)
この本文は AI(Claude)が読むための原文(英語または中国語)です。日本語訳は順次追加中。
Val Town
Overview
Val Town is a platform for writing and deploying TypeScript functions instantly — no infrastructure, no build step, no deployment pipeline. Write a function in the browser, get a URL. HTTP endpoints, cron jobs, email handlers, and persistent SQLite storage. Think "GitHub Gists that run."
When to Use
- Need a quick API endpoint or webhook handler (minutes, not hours)
- Scheduled tasks (cron) without managing servers
- Prototyping an idea before building proper infrastructure
- Webhook receivers for Stripe, GitHub, Slack integrations
- Glue code between services (fetch from API A, transform, POST to API B)
- Storing small amounts of data with built-in SQLite
Instructions
HTTP Val (API Endpoint)
// @user/myApi — Deployed instantly at https://user-myapi.web.val.run
export default async function(req: Request): Promise<Response> {
const url = new URL(req.url);
if (req.method === "GET") {
const name = url.searchParams.get("name") || "World";
return Response.json({ message: `Hello, ${name}!` });
}
if (req.method === "POST") {
const body = await req.json();
// Process the data
return Response.json({ received: body, timestamp: Date.now() });
}
return new Response("Method not allowed", { status: 405 });
}
Cron Val (Scheduled Task)
// @user/dailyReport — Runs on a schedule
export default async function() {
// Fetch data from an API
const response = await fetch("https://api.example.com/stats");
const stats = await response.json();
// Send to Slack
await fetch(Deno.env.get("SLACK_WEBHOOK")!, {
method: "POST",
body: JSON.stringify({
text: `📊 Daily Report: ${stats.users} users, ${stats.revenue} revenue`,
}),
});
}
SQLite Storage
// @user/todoApi — CRUD API with persistent SQLite storage
import { sqlite } from "https://esm.town/v/std/sqlite";
// Initialize table
await sqlite.execute(`
CREATE TABLE IF NOT EXISTS todos (
id INTEGER PRIMARY KEY AUTOINCREMENT,
title TEXT NOT NULL,
done BOOLEAN DEFAULT FALSE,
created_at DATETIME DEFAULT CURRENT_TIMESTAMP
)
`);
export default async function(req: Request): Promise<Response> {
const url = new URL(req.url);
if (req.method === "GET") {
const todos = await sqlite.execute("SELECT * FROM todos ORDER BY created_at DESC");
return Response.json(todos.rows);
}
if (req.method === "POST") {
const { title } = await req.json();
await sqlite.execute("INSERT INTO todos (title) VALUES (?)", [title]);
return Response.json({ ok: true }, { status: 201 });
}
if (req.method === "DELETE") {
const id = url.searchParams.get("id");
await sqlite.execute("DELETE FROM todos WHERE id = ?", [id]);
return Response.json({ ok: true });
}
return new Response("Not found", { status: 404 });
}
Webhook Handler
// @user/stripeWebhook — Handle Stripe webhooks
export default async function(req: Request): Promise<Response> {
const signature = req.headers.get("stripe-signature");
const body = await req.text();
// Verify webhook signature
// In Val Town, use Deno.env.get() for secrets
const secret = Deno.env.get("STRIPE_WEBHOOK_SECRET");
const event = JSON.parse(body);
switch (event.type) {
case "checkout.session.completed":
// Handle successful payment
await fetch("https://api.myapp.com/activate", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ customerId: event.data.object.customer }),
});
break;
case "customer.subscription.deleted":
// Handle cancellation
break;
}
return Response.json({ received: true });
}
Examples
Example 1: Quick monitoring endpoint
User prompt: "I need a quick URL that checks if my website is up and returns the status."
The agent will create an HTTP val that fetches the target URL, measures response time, and returns a JSON status report.
Example 2: GitHub webhook to Slack
User prompt: "When someone stars my GitHub repo, send a message to my Slack channel."
The agent will create an HTTP val that handles GitHub webhook events, filters for star events, and posts to a Slack webhook URL.
Guidelines
- HTTP vals are standard Web API —
Requestin,Responseout - Environment variables via
Deno.env.get()— store secrets in Val Town settings - SQLite is per-account — shared across all your vals, persistent
- Free tier: 10 vals, 100 cron runs/day — enough for prototyping
- Import from URLs —
import { x } from "https://esm.town/v/user/module" - Deno runtime — use Deno APIs, npm packages via
npm:packagespecifier - No cold starts — vals are always warm, sub-50ms response times
- Use for glue code — connect APIs, transform data, automate workflows
- Not for production traffic — great for webhooks, cron, prototypes; use proper infra for high-traffic APIs