# Web Browsing

> Fetch and read web pages using command-line tools

- Skill: `pinpox/web-browsing` (Agent Skill)
- Install (CLI): `npx skillmds@latest add pinpox/web-browsing`
- Raw SKILL.md: https://api.skillmd.com/api/skills/pinpox/web-browsing/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: pinpox (https://skillmd.com/u/pinpox)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/pinpox/web-browsing

---


You can browse the web using command-line tools via the bash tool.

### Quick fetch (raw HTML or API responses)

```bash
curl -sL "https://example.com"
```

Use `-sL` to follow redirects silently. Add `-o /dev/null -w "%{http_code}"` to check status codes.

### Rendered text (readable output)

```bash
lynx -dump "https://example.com"
```

Or with w3m:

```bash
w3m -dump "https://example.com"
```

These render HTML to plain text, making web pages readable without HTML tags.

### Downloading files

```bash
curl -sLO "https://example.com/file.tar.gz"
```

### JSON APIs

```bash
curl -s "https://api.example.com/data" | jq .
```

Use `jq` to parse and filter JSON responses.

### Tips

- Always use `curl -sL` (silent + follow redirects) as the default.
- For long pages, pipe through `head -n 100` to avoid excessive output.
- Use `lynx -dump` when you need to read a web page — it handles tables, links, and formatting.
- Check if `lynx` or `w3m` are available with `which lynx w3m` before using them.