Scrapeless Cline

Give Cline live web access — search, scrape, screenshot, and a real cloud browser — by adding the Scrapeless MCP server. 21 tools, one settings block.

Use case

Cline edits your repo autonomously but cannot see the live web. Adding this server lets it:

• Write a scraper against the real page — read the markup, propose selectors, verify them, and correct itself before handing you the file.
• Read current docs for a library it is integrating, instead of relying on training data.
• Collect data into your repo — SERPs, competitor pages, product listings — as files you can commit.
• Check a deployed page with a screenshot and describe what rendered.

Requirements

• VS Code with the Cline extension
• Node.js 18+ (for npx)
• A Scrapeless API key — create a free account

Setup

Cline keeps MCP servers in cline_mcp_settings.json, which lives in VS Code's global storage — outside your project. Open it the safe way: click the MCP Servers icon in the Cline panel → Configure MCP Servers. That opens the exact file for your install.

Or edit it directly:

OS	Path
macOS	~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
Linux	~/.config/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
Windows	%APPDATA%\Code\User\globalStorage\saoudrizwan.claude-dev\settings\cline_mcp_settings.json

Paste the block from cline_mcp_settings.example.json and replace the placeholder with your key:

{
  "mcpServers": {
    "scrapeless": {
      "command": "npx",
      "args": ["-y", "scrapeless-mcp-server"],
      "env": { "SCRAPELESS_KEY": "your_key_here" },
      "disabled": false,
      "autoApprove": ["google_search", "scrape_markdown", "scrape_html"]
    }
  }
}

Because this file sits outside your repository, the key in it is never at risk of being committed — put the literal value there rather than relying on shell-variable expansion.

autoApprove lists the read-only tools that run without a confirmation prompt each time. The three above only fetch data. Leave the browser_* tools out of that list so session-creating and clicking actions stay manual.

The one thing that will bite you

The server reads SCRAPELESS_KEY, not SCRAPELESS_API_KEY. Verified behavior:

Environment	Result
SCRAPELESS_KEY set	Works — tools respond
Only SCRAPELESS_API_KEY set	Server starts, then never responds — no error, just a hang
Neither set	Server starts, then never responds

No credential error is raised — the tool call just hangs, which in Cline looks like the task freezing mid-step. Use SCRAPELESS_KEY in the settings file.

Verify before you trust it

verify_mcp.py speaks MCP over stdio — the same protocol Cline uses:

export SCRAPELESS_API_KEY=your_key_here    # this script accepts either name
python3 verify_mcp.py           # list the tools
python3 verify_mcp.py --call    # also scrape example.com

Real output:

server: scrapeless-mcp-server 0.2.0
tools : 21
...
calling scrape_markdown on https://example.com ...
returned 184 characters of markdown

If this passes and Cline still shows the server as failed, the problem is the settings file — check for trailing commas and that the JSON parses.

The 21 tools

Search and one-shot fetch — stateless: google_search, google_trends, scrape_html, scrape_markdown, scrape_screenshot.

Cloud browser session — stateful; browser_create first, browser_close last: browser_goto, browser_get_html, browser_get_text, browser_snapshot, browser_screenshot, browser_click, browser_type, browser_press_key, browser_scroll, browser_scroll_to, browser_wait, browser_wait_for, browser_go_back, browser_go_forward.

Full list with input schemas: results/tools.json.

Prompts that work

Use scrape_html on <url>, show me the markup around the listing cards, propose
selectors for title and price, then write scraper.py and re-check the selectors
with scrape_html before you finish.

Use scrape_markdown on <docs url> and write an integration that matches the
current API — do not rely on your training data for the parameter names.

Troubleshooting

Symptom	Cause and fix
Task freezes on a tool call	SCRAPELESS_KEY missing. See the table above.
Server not listed	JSON is malformed, or you edited the wrong file. Use Configure MCP Servers from the Cline panel.
Approval prompt on every fetch	Add the read-only tools to autoApprove.
npx re-downloads each launch	npm install -g scrapeless-mcp-server, then set "command": "scrapeless-mcp-server" with no args.
Browser tools error about no session	Call browser_create first.
Nearly empty page content	Read before the challenge cleared. browser_wait_for a content selector, then read.

Project structure

scrapeless-cline/
├── cline_mcp_settings.example.json   # paste into Cline's settings file
├── verify_mcp.py                     # stdio MCP self-check, no editor required
├── results/tools.json                # live tool list with input schemas
└── LICENSE

Related

• Product: Scraping Browser · Universal Scraping API
• Integration page: Scrapeless with Cline
• Upstream server: scrapeless-ai/scrapeless-mcp-server

License

MIT — see LICENSE.