技能详情（站内镜像，无评论）
许可证：MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本：v1.0.0
统计：⭐ 1 · 30 · 0 current installs · 0 all-time installs
⭐ 1
安装量（当前） 0
🛡 VirusTotal ：良性 · OpenClaw ：良性
Package：claudy-my-laudy/jina-ai-skill
安全扫描（ClawHub）
- VirusTotal ：良性
- OpenClaw ：良性
OpenClaw 评估
The skill does what it says: small Python helpers and curl examples that call r.jina.ai and s.jina.ai to return markdown/search results; nothing requested or installed is disproportionate to that purpose.
目的
Name and description match the provided scripts and SKILL.md. The two included Python scripts only call the r.jina.ai and s.jina.ai endpoints and optionally read a JINA_API_KEY env var for higher rate limits—this is proportional to a web-reading/search skill.
说明范围
Instructions and scripts only perform outbound HTTP requests to Jina's reader/search endpoints and print results. This is within scope. Important privacy note: any URL or content you send (or that the skill instructs the remote service to fetch) will be transmitted to jina.ai; avoid passing secrets or private/internal URLs you don't want sent to a third party.
安装机制
No install spec; this is instruction-plus-small-scripts only. The Python files use the standard library (urllib) and do not download additional code or third-party packages.
证书
No required environment variables are declared. An optional JINA_API_KEY may be provided for higher rate limits; this is appropriate and limited in scope.
持久
Skill does not request persistent/always-on privileges, does not modify other skills or system config, and defaults to normal agent invocation behavior.
综合结论
This skill is coherent and small, but be aware of two practical risks before installing: (1) Privacy/exfiltration — when you ask it to 'read' a URL the URL (and thus whatever the remote service fetches) is sent to Jina's servers (r.jina.ai/s.jina.ai). Do not pass private, internal, or secret URLs or content you don't want a third party to see. (2) Input validation — the scripts do minimal validation; if used by an automated agent, ensure it wo…
安装（复制给龙虾 AI）
将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「Jina Reader」。简介：Fetch the markdown content of any webpage using Jina AI's Reader API (r.jina.ai…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/claudy-my-laudy/jina-ai-skill/SKILL.md
（来源：yingzhi8.cn 技能库）SKILL.md
---
name: jina-reader
description: "Fetch the markdown content of any webpage using Jina AI's Reader API (r.jina.ai), or search the web using Jina AI's Search API (s.jina.ai). Use this skill whenever you need to read a URL and get clean LLM-friendly markdown output, extract article/documentation content from a page, or search the web and retrieve rich results with content. Triggers on: read this URL, fetch this page, get the content of, what does this page say, summarize this link, search the web for, look up, or any request requiring clean markdown from a URL."
---
# Jina Reader & Search Skill
Two APIs, zero API key required (optional key for higher rate limits).
## Reader — Fetch a URL as Markdown
**Simple (shell):**
```bash
curl -s "https://r.jina.ai/<URL>"
```
**With script:**
```bash
python3 skills/jina-reader/scripts/jina_read.py <URL>
python3 skills/jina-reader/scripts/jina_read.py <URL> --no-images # strip images
python3 skills/jina-reader/scripts/jina_read.py <URL> --json # structured JSON response
```
**With API key (set in .env):**
```bash
JINA_API_KEY=your_key python3 skills/jina-reader/scripts/jina_read.py <URL>
```
**Useful request headers (for direct curl):**
| Header | Purpose |
|--------|---------|
| `Authorization: Bearer <key>` | Higher rate limits |
| `X-Return-Format: markdown` | Force markdown output |
| `X-Remove-Selector: img` | Strip images |
| `Accept: application/json` | JSON response with title, url, content |
## Search — Web Search via Jina
**Simple (shell):**
```bash
curl -s "https://s.jina.ai/<URL-encoded-query>"
```
**With script:**
```bash
python3 skills/jina-reader/scripts/jina_search.py "your query"
python3 skills/jina-reader/scripts/jina_search.py "your query" --json
python3 skills/jina-reader/scripts/jina_search.py "your query" --json --results 3
```
## API Key
| API | Without Key | With Key |
|-----|-------------|----------|
| Reader (`r.jina.ai`) | ✅ Works (rate-limited) | Higher limits |
| Search (`s.jina.ai`) | ❌ Requires key | ✅ Works |
Store as `JINA_API_KEY` in `.env`. Get a free key at: https://jina.ai/api-dashboard
> **No key for search?** Use the `duckduckgo-search` skill as a fallback.
## Notes
- Reader works on most pages including JS-heavy sites
- Search returns up to 5 results by default, each with title, URL, and content summary
- For very long pages, use `--no-images` to reduce token usage