mcpbeat

Harvest Single

vibeeval/harvest-single

Single page smart extraction - articles, docs, blog posts to clean markdown

392 tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
521
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository 
npx skills add https://github.com/vibeeval/vibecosystem --skill harvest-single

What it tells the agent to use

found in the instruction text
WebFetch fetches pages from the network

The instruction itself

7 sections, as written by the author

Harvest Single Page

Extract and clean content from a single web page. Auto-detects content type (article, documentation, API reference, blog post) and produces clean, structured markdown.

Usage

/harvest <url>

Examples

# Extract a blog post
/harvest https://blog.example.com/best-practices-2024

# Extract API documentation page
/harvest https://docs.stripe.com/api/charges

# Extract a GitHub README
/harvest https://github.com/owner/repo

How It Works

  • Fetch URL content via WebFetch or crawl4ai
  • Detect content type (article, docs, API ref, blog, wiki)
  • Extract main content, strip navigation/ads/footers
  • Preserve code blocks, tables, images
  • Add metadata header (source, date, word count)
  • Save to .claude/cache/agents/harvest/

Output Format

# [Page Title]
> Source: [URL]
> Extracted: [timestamp]
> Type: [article|docs|api|blog|wiki]
> Words: [count]

[Clean extracted content in markdown]

## Links Found
- [Link text](URL)

Fallback Chain

  • crawl4ai Docker (port 11235) - preferred
  • WebFetch tool - built-in fallback
  • curl + html2text - last resort

When to Use

  • Quick grab of a single page's content
  • Extracting a specific doc page for reference
  • Saving an article for later analysis
  • Getting clean markdown from messy HTML

How to use it

Copy the folder

Take vibeeval/harvest-single from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.