Safe Smart Web Fetch

安全网页抓取技能。获取网页内容时，默认先判断 URL 是否可能包含 token、是否为内网/本地域名、是否为私密链接；这三类一律不走第三方清洗服务，只走直接抓取。其余公开网页可按顺序尝试 Jina Reader、markdown.new、defuddle.md 获取干净 Markdown，失败再回退原始抓取。

This skill appears coherent and does what it says: it classifies URLs and, for public pages, tries third‑party cleaners before falling back to original fetch. Before installing or running it against sensitive content, consider: 1) review and, if necessary, replace or restrict the third‑party endpoints (r.jina.ai, markdown.new, defuddle.md) to services you trust or to an internal sanitizer; 2) test the classifier against representative internal/private URLs to confirm it blocks them from third‑party calls (heuristics may miss uncommon token names or edge hostnames); and 3) if you need stronger guarantees, add explicit allow/deny lists or require explicit user confirmation before sending any URL containing query parameters or single‑label hosts to external services.

Type: OpenClaw Skill Name: safe-smart-web-fetch Version: 1.0.0 The skill provides a utility for fetching web content while implementing privacy safeguards. The script `scripts/fetch.py` includes logic to detect sensitive URLs (containing tokens, session IDs, or administrative paths) and private network addresses (localhost, private IP ranges, internal hostnames), ensuring these are fetched directly rather than being sent to third-party Markdown conversion services like Jina Reader (r.jina.ai), markdown.new, or defuddle.md. The implementation uses standard Python libraries and aligns perfectly with the safety-oriented instructions in `SKILL.md`.