# Scrape Page

> Scrape a web page into clean, agent-readable content with Firecrawl. Use when the agent needs the full content of a specific page (docs, articles, reference pages) for reading, extraction, or summarization.

- Skill: `hiai-gg/scrape-page` (Agent Skill)
- Install (CLI): `npx skillmds@latest add hiai-gg/scrape-page`
- Raw SKILL.md: https://api.skillmd.com/api/skills/hiai-gg/scrape-page/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: HiAi-gg (https://skillmd.com/u/hiai-gg)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/hiai-gg/scrape-page

---


# Scrape a Page with Firecrawl

Use this skill when an agent needs the clean content of a specific URL.

## Workflow

### 1. Confirm the target

- Get the exact URL from the user or from a search result.
- State what you expect to extract (full text, specific section, structured
  fields) so you can verify the output.

### 2. Scrape

```text
firecrawl_scrape(url)
```

The result is normalized, agent-readable content (markdown by default).

### 3. Verify the extraction

- Check the content actually covers the target page (correct page, no captcha
  wall, no empty body).
- If the page is interactive-heavy (JS-rendered SPA), note that the scrape may
  miss content; prefer a static URL or use browser interaction tools where
  available.
- If you need only part of the page, extract that part rather than dumping the
  whole body into the response.

### 4. Report

Give: the source URL, a summary of the content (or the requested excerpt), and
any caveats (paywall, partial render, robots-limited). Do not include the full
page text unless the user asked for it.

## Guardrails

- Scrape only URLs the user provided or that came from the user's task.
- Respect site terms and robots directives; do not bypass paywalls or access
  controls.
- Never pass credentials through the scrape call.