mcpField
J
jina-reader

jina-reader

github.com/dwsyProductivity2440 installsUpdated 5w ago
Repository

About jina-reader

jina-reader is a skill for extracting URL content and searching the web. It suits workflows that need webpage material, with multiple output formats and CSS-selector filtering.

Key features

  1. 01Read URLs as Markdown
  2. 02Output HTML, text, or JSON
  3. 03Extract content with CSS selectors
  4. 04Search and filter by site

Use cases

  1. 01Extract webpage text for later processing
  2. 02Search a topic or specific site

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/dwsy/agent --skill jina-reader

After installing, type/jina-readerin Claude Code to confirm it loaded.

Bundled filesbun.lockscripts/reader.ts

Details

From the project README

Jina Reader

URL 内容提取 + 网络搜索。

执行

# 读取 URL（转 Markdown）
bun ~/.pi/agent/skills/jina-reader/scripts/reader.ts read "<url>"

# 搜索
bun ~/.pi/agent/skills/jina-reader/scripts/reader.ts search "<query>"

# 选项
read <url> [--format markdown|html|text|json] [--timeout N] [--no-cache] [--selector ".class"]
search <query> [--count N] [--site domain.com] [--type web|images|news]

API Key

~/.pi/agent/skills/jina-reader/.env 中配置 JINA_API_KEY。

示例

bun scripts/reader.ts read "https://docs.jina.ai"
bun scripts/reader.ts search "AI news" --count 5 --site arxiv.org
bun scripts/reader.ts read "https://example.com" --selector ".article" --format json

Comments

Markdown supported