parallel-web-extract

使用Parallel AI REST API（curl）从URL中抓取并提取结构化内容。适用于价格页面、文档、产品信息等。无需二进制安装 —— 需要在.env.local文件中设置PARALLEL_API_KEY。

Parallel Web Extract

Extract structured content from any URL using Parallel AI. Results in 5-30 seconds.

CLI alternative (recommended): Install parallel-cli for official skill: npx skills add parallel-web/parallel-agent-skills --skill parallel-web-extract

Setup

API_KEY=$(grep "^PARALLEL_API_KEY=" .env.local | cut -d= -f2)

Extract Request

curl -s -X POST "https://api.parallel.ai/v1beta/extract" \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -H "parallel-beta: search-extract-2025-10-10" \
  -d '{
    "url": "https://example.com/pricing",
    "objective": "Extract all pricing plans and their features"
  }'

When to Use

Scrape a specific URL you already know
Extract pricing, specs, or structured data from a page
Get clean content from documentation sites

For finding URLs first, use parallel-web-search then extract.

Error Handling

| Code | Meaning | Fix | |------|---------|-----| | 401 | Bad key | Check PARALLEL_API_KEY in .env.local | | 400 | Invalid URL | Ensure URL is accessible | | 429 | Rate limit | Wait 60s |