Extracting web data

Skill Scout-AI-Labs/scout-skills/skills/extracting-web-data

Agent Skills for using Scout: web research, sourced answers, company enrichment, structured extraction, monitoring. Works in Claude Code, Claude.ai, the Agent SDK, and Vercel eve.

Install

npx -y skills add Scout-AI-Labs/scout-skills --skill extracting-web-data

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Pull structured data from web pages with Scout. Use when the user wants specific fields out of one or more URLs (prices, specs, contact details, table rows, product attributes) rather than the whole page, or wants the result shaped to a JSON schema. Give Scout the URLs and an objective and it returns structured JSON.

SKILL.md

1.6 KB, 329 tokens by cl100k_base, as published. Nobody here has run it

Extracting structured web data

When the user wants named fields, not the whole page, use extract instead of scrape. Describe what you want as an objective, or pass a JSON Schema for an exact shape.

curl -s https://core.usescout.sh/v1/extract \
  -H "Authorization: Bearer $SCOUT_API_KEY" -H "Content-Type: application/json" \
  -d '{
    "urls": ["https://example.com/pricing"],
    "objective": "plan names, monthly price, and the included seat count"
  }'

For a guaranteed shape, send output_schema (a JSON Schema object) instead of, or alongside, the objective:

{
  "urls": ["https://example.com/pricing"],
  "output_schema": {
    "type": "object",
    "properties": { "plans": { "type": "array", "items": {
      "type": "object",
      "properties": { "name": {"type":"string"}, "price": {"type":"number"} }
    } } }
  }
}

With the MCP server, use scout_extract. To find pages first when you only have a description, set "find_via_search": true with "search_queries".

Notes

For a single page where you want the readable text, scrape to Markdown is simpler. Reach for extract when you need fields.
Extract several URLs in one call by passing them all in urls; the result keys back to each source.