AgentSkillsCN

scratchpad-fetch

下载并聚合网页/文档到带时间戳的草稿文件。在用户要求“合并所有这些资源”、“获取所有这些链接”、“查看所有这些资源”或想从文档URL中收集新鲜上下文时使用。一个提示的所有URL都会进入docs/scratchpad/<timestamp>.md的单个文件。

SKILL.md
--- frontmatter
name: scratchpad-fetch
description: Download and aggregate web pages/docs into timestamped scratchpad files. Use when user asks to "concatenate all these resources", "get all these links", "checkout all these resources", or wants to gather fresh context from documentation URLs. All URLs from one prompt go into single file at docs/scratchpad/<timestamp>.md.
allowed-tools: Bash

Scratchpad Fetch

Overview

Downloads web pages via curl and appends content to timestamped scratchpad file. All URLs from single user prompt → single file. Simple context gatherer for documentation.

Usage

When user provides URLs and asks to:

  • "concatenate all these resources"
  • "get all these links"
  • "checkout all these resources"
  • "gather these docs"
  • "download these pages"

Execute:

bash
./skills/scratchpad-fetch/scripts/fetch_urls.sh <url1> <url2> <url3> ...

Output: docs/scratchpad/YYYYMMDD_HHMMSS.md

Script Behavior

fetch_urls.sh:

  • Creates docs/scratchpad/ if missing
  • Generates timestamp: YYYYMMDD_HHMMSS
  • Downloads each URL with curl
  • Appends all content to single file
  • Adds headers/separators for readability
  • Reports success/failure per URL

Example

User: "checkout all these resources: https://example.com/api https://example.com/guide"

bash
./skills/scratchpad-fetch/scripts/fetch_urls.sh \
  https://example.com/api \
  https://example.com/guide

Output: docs/scratchpad/20250129_143052.md containing both pages.

Notes

  • Dumb but effective for gathering fresh docs
  • One prompt = one file
  • curl follows redirects (-L)
  • Failures logged but don't stop other URLs