Crawlberg MCP
Crawlberg is an open-source web crawling and ingestion tool that converts websites into clean, structured Markdown, metadata, and links. It handles JavaScript-heavy pages via headless browser fallback, respects robots.txt and sitemaps, and provides a single Rust core with bindings for 14 languages. It is designed to feed structured data into AI pipelines and agents.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Crawl a single URL to extract Markdown and metadata
- Map entire websites by following links to structured data
- Integrate web scraping into AI agent workflows via MCP
- Extract social-card and JSON-LD data from pages
Source and maintenance
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
ghcr.io/kreuzberg-dev/kreuzcrawl:0.3.0
{
"transport": "stdio",
"identifier": "ghcr.io/kreuzberg-dev/kreuzcrawl:0.3.0",
"registryType": "oci",
"packageArguments": [
{
"type": "positional",
"value": "mcp"
}
]
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"crawlberg-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kreuzberg-dev/kreuzcrawl:0.3.0",
"mcp"
]
}
}
}Cline
{
"mcpServers": {
"crawlberg-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kreuzberg-dev/kreuzcrawl:0.3.0",
"mcp"
]
}
}
}VS Code
{
"servers": {
"crawlberg-mcp": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kreuzberg-dev/kreuzcrawl:0.3.0",
"mcp"
]
}
}
}Claude Desktop
{
"mcpServers": {
"crawlberg-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kreuzberg-dev/kreuzcrawl:0.3.0",
"mcp"
]
}
}
}Let your agent find its tools
Find servers like Crawlberg MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.