# Extract Web Data with Just Scrape

Manual web research and scraping can be slow and inconsistent. This skill guides Claude, Codex, and Claude Code through ScrapeGraph CLI workflows.

## Install

```bash
npx skillstore add scrapegraphai/just-scrape
```

## Metadata

- Status: approved
- Slug: scrapegraphai-just-scrape
- Skillstore revision: r1
- Version status: missing
- Tree hash: 22f3f5c46412b7eea2f7f88c631bd3c18512568641a71255b4e09d775f2c682e
- Author: scrapegraphai
- GitHub username: scrapegraphai
- License: MIT
- Repository: https://github.com/scrapegraphai/just-scrape/tree/main/skills/just-scrape/
- Ref: 62e2a730c5cd74eab4c7164309d810de660fcea3
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network, filesystem, env\_access
- Quality score: 70
- Public page: https://skillstore.pages.dev/skills/scrapegraphai-just-scrape
- Manifest: https://skillstore.pages.dev/api/skills/scrapegraphai-just-scrape/manifest

## Capabilities

- Search the web with result limits, country filters, time ranges, and optional structured extraction.
- Scrape known URLs into markdown, HTML, screenshots, links, images, summaries, branding, or JSON.
- Extract structured fields from pages using prompts and optional schemas.
- Crawl bounded site sections with page limits, depth limits, and include or exclude patterns.
- Create, inspect, update, pause, resume, and delete page monitors.
- Check request history, credit balance, and API setup with CLI commands.

## Use Cases

- Research Competitor Pages: Find public pages, scrape visible content, and summarize pricing, positioning, or feature changes.
- Collect Documentation Content: Crawl a bounded documentation section and save markdown or links for review, migration, or support answers.
- Monitor Page Changes: Create recurring checks for public pages and inspect monitor activity when content changes.

## Prompt Templates

### Scrape One Page

```
Use just-scrape to scrape this URL as markdown and summarize the main points. Save the raw output under .just-scrape.
```

### Search Before Scraping

```
Search the web for recent sources about this topic. Return the best sources with titles, URLs, and a short reason for each.
```

### Extract Structured Fields

```
Extract product names, prices, availability, and source URLs from this page. Use a strict schema and note missing fields.
```

### Crawl and Monitor Safely

```
Crawl this documentation section with clear page and depth limits. Then suggest a monitor plan for the pages most likely to change.
```

## Limitations

- Requires the just-scrape CLI and a valid ScrapeGraph AI API key.
- Consumes ScrapeGraph credits for searches, scrapes, crawls, and monitors.
- Results depend on website access, robots policies, JavaScript rendering, and page stability.
- Sensitive headers, cookies, and scraped content need careful handling outside the skill.

## Best Practices

- Confirm the target URL and data fields before running high-credit operations.
- Set crawl limits and include patterns before scraping site sections.
- Store outputs in .just-scrape and review large files incrementally.

## Anti Patterns

- Do not scrape protected pages with session cookies unless the user owns the access.
- Do not run broad crawls or many monitors without explicit limits.
- Do not paste API keys, bearer tokens, or cookies into prompts or logs.

## Security Audit

- Audited at: 2026-07-05T21:02:08.454\+00:00
- Summary: The static findings are mostly false positives caused by Markdown backticks, documented CLI commands, expected API key names, and example URLs. No prompt injection or covert credential exfiltration was found in SKILL.md. The remaining concern is an unpinned global npm install path for a community CLI package.

## Stats

- Views: 87
- Downloads: 12
- Favorites: 0
- Popularity score: 0