domscout
Browser API as agent tools: Markdown, screenshots, structured data, automation, crawls
- Pricing
- Freemium
- Platforms
- CLI, API, Web
- Runs
- Locally
- Version
- 1.0.8
- Listing updated
- 1 day ago
About domscout
domscout is a freemium MCP server for web search and scraping. It runs locally on your machine and is started by your AI client.
domscout turns its browser API into ten MCP tools. An agent can read a page as clean Markdown, capture a PNG, JPEG, WebP or PDF, extract named fields as typed JSON, read a page's accessibility tree, run allowed page actions, or start a bounded crawl. Long jobs are durable, so the agent can poll or cancel them, and completed screenshots come back as MCP image blocks.
Every tool has a credit price and no request costs more than 10 credits. Results report what the call cost and what remains, so an agent can budget without another call, and domscout_check_credits shows which capabilities the key can use. Crawls only run on allowlisted sites and respect robots.txt. The docs add that web search is deliberately not wrapped while no provider is configured.
There are two ways in: a hosted endpoint at domscout.io/mcp with OAuth or a bearer key, or the MIT-licensed npm package @domscout/mcp run through npx with Node 20 or newer. Without a key, the Markdown, screenshot and credit tools run on a keyless demo of a few calls an hour. The server holds no browser, database or billing credential.
Install
Run in a terminal
claude mcp add domscout --env DOMSCOUT_API_KEY='<DOMSCOUT_API_KEY>' -- npx -y @domscout/mcpAdd to claude_desktop_config.json
{
"mcpServers": {
"domscout": {
"command": "npx",
"args": [
"-y",
"@domscout/mcp"
],
"env": {
"DOMSCOUT_API_KEY": "<DOMSCOUT_API_KEY>"
}
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"domscout": {
"command": "npx",
"args": [
"-y",
"@domscout/mcp"
],
"env": {
"DOMSCOUT_API_KEY": "<DOMSCOUT_API_KEY>"
}
}
}
}Add to .vscode/mcp.json
{
"servers": {
"domscout": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@domscout/mcp"
],
"env": {
"DOMSCOUT_API_KEY": "<DOMSCOUT_API_KEY>"
}
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"domscout": {
"command": "npx",
"args": [
"-y",
"@domscout/mcp"
],
"env": {
"DOMSCOUT_API_KEY": "<DOMSCOUT_API_KEY>"
}
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"domscout": {
"command": "npx",
"args": [
"-y",
"@domscout/mcp"
],
"env": {
"DOMSCOUT_API_KEY": "<DOMSCOUT_API_KEY>"
}
}
}
}Configuration
DOMSCOUT_API_KEYA domscout API key from https://www.domscout.io/dashboard/api-keys. Every tool call is billed to the account that owns it.DOMSCOUT_BASE_URLOverride the API host. Only needed against a non-production deployment; omit it in normal use.DOMSCOUT_DOCS_BASE_URLOverride the host the documentation resources are read from. It cannot be derived from DOMSCOUT_BASE_URL, because the API host serves no llms-full.txt.
Tools this server provides (10)
Taken from the project documentation; the running server may differ.
- domscout_check_credits
- Shows balance, quota, rate limit and price list.
- domscout_extract_markdown
- Reads a page as clean Markdown.
- domscout_capture_screenshot
- Captures PNG, JPEG, WebP or PDF.
- domscout_extract_data
- Extracts named fields as typed JSON.
- domscout_inspect_page
- Reads the accessibility tree and interactive elements.
- domscout_automate_page
- Runs allowed actions and captures the result.
- domscout_crawl_site
- Runs a bounded, allowlisted, robots-respecting crawl.
- domscout_get_job
- Polls a durable job, batch or crawl.
- domscout_cancel_job
- Cancels a durable job.
- domscout_send_feedback
- Sends product feedback.
What it does
- Page to Markdown
- Reads a page as clean Markdown, with a fast mode that skips the browser.
- Screenshots and PDFs
- Captures PNG, JPEG, WebP or PDF at 1 to 3 credits.
- Typed data extraction
- Pulls named fields out of a page as JSON.
- Page inspection
- Reads the accessibility tree and interactive elements.
- Page automation
- Runs allowed actions and captures the outcome.
- Bounded crawls
- Crawls allowlisted sites while respecting robots.txt.
- Credit reporting
- Each result shows its cost and the remaining balance.
Good for
- Have an agent read a web page as Markdown
- Capture a screenshot of a page for review
- Extract product or listing fields into JSON
- Inspect a page's interactive elements before automating it
- Crawl a documentation site within set limits
Pricing
Freemium
Credit-based; free key available, plan gates apply; prices not listed here
Prices change. Check domscout.io before you buy.
domscout FAQ
Do I need an API key?
For most tools, yes; without one, the Markdown, screenshot and credit tools run on a small keyless demo.
What is the maximum cost of a request?
No request exceeds 10 credits.
Does it respect robots.txt?
Yes, the crawl tool is allowlisted and robots-respecting.
Can it run web searches?
No. The search route is withheld while no provider is configured, so no tool wraps it.
Is domscout free?
domscout has a free plan, with paid plans for more usage or features.
How do I add domscout to Claude Code?
Run this command in your terminal: claude mcp add domscout --env DOMSCOUT_API_KEY='<DOMSCOUT_API_KEY>' -- npx -y @domscout/mcp
What are the best alternatives to domscout?
The closest alternatives to domscout are Browser Use, Scrapling MCP Server, Firecrawl MCP Server, SearXNG Search and Oya Browser.
Alternatives to domscout

Browser Use
MCP serverBrowser agents and cloud browsers for AI, with an MCP server for web tasks
- 7 tools
- Remote
- Freemium
- 121K GitHub stars
- Browser automation

Scrapling MCP Server
MCP serverMCP server for AI clients to scrape sites with HTTP and stealth browser sessions
Runs locally- 17 tools
- Open source
- 104K GitHub stars
- Web search and scraping

Firecrawl MCP Server
MCP serverMCP server that lets AI agents search, scrape and interact with web pages
- 3 tools
- Remote
- Freemium
- 7.8K GitHub stars
- Web search and scraping

SearXNG Search
MCP serverMCP server for web search and page reading through a SearXNG instance
Runs locally- 3 tools
- Open source
- 1.4K GitHub stars
- Web search and scraping

Oya Browser
MCP serverHIPAA-oriented agentic browser with replayable playbooks, reachable over MCP
- 32 tools
- Remote
- Freemium
- 349 GitHub stars
- Browser automation
Mantis
MCP serverBrowser MCP server that gives agents numbered text pages instead of screenshots
Runs locally- 1 tool
- Open source
- 5 Hacker News points
- Browser automation