Olostep MCP Server
未认领MCP server for Olostep — the web scraping, crawling, and search infrastructure used by top AI companies. Gives any MCP-compatible AI agent the ability to scrape, crawl, batch-extract, and search the web in real time.
安装
docker run -i --rm -e OLOSTEP_API_KEY="your-api-key" olostep/mcp-server:local工具（17 个）
answer_id
, object, task, result (JSON if provided), sources, created
batch_id
, status, totalurls, createdat, formats, country, parser, urls
country
Optional country code (e.g., US, GB, CA) for location-specific scraping
crawl_id
, object, status, starturl, maxpages, created, formats, country, parser
cursor
Pagination cursor into the list of discovered pages (default: 0)
formats
Array of formats to retrieve per page — markdown, html, json, text (default: ["markdown"])
items_limit
Max pages to retrieve content for, 1–100 (default: 20)
json
Optional JSON schema/object or a short description of the desired output shape
map_id
, object, url, totalurls, urls, searchquery, topn
output_format
Choose format (html, markdown, json, or text) - default: markdown
parser
Optional parser ID for specialized extraction
query
Search query (required)
search_query
The search query to sort URLs by (required)
task
Question or task to answer using web data (required)
url
The URL of the website to map (required)
url_to_scrape
The URL of the website you want to scrape (required)
wait_before_scraping
Wait time in milliseconds before scraping (0-10000)