Server facts
scrapi-mcp-server
Project
- Author
- bamchi
- Repository
- github.com/bamchi/scrapi-mcp-server
- License
- MIT
- Category
- Search & Web Data
Tech
- Runtime / language
- node
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 5
- Last commit
- Data as of
Tags
- Markdown
- Streamable Http
- Web Scraping
- ai agent
- anti bot
- llm
- mcp
- scraping
Overview
Web content retrieval in Scrapi centers on two tools: scrape_url for individual pages and scrape_urls for parallel requests covering up to 10 URLs. Both offer Markdown or plain text, with Markdown as the default. Extraction retains document structure such as headings, lists, and links while removing advertising, navigation, and scripts. Metadata can include page titles, authors, dates, and images. Browser rendering supports JavaScript applications, giving agents source material for workflows such as summarizing articles, checking competitor prices, or reading API documentation before writing integration code.
Operational checks sit alongside content extraction. get_usage reports the current plan and separates subscription credits from purchased credits and the remaining total. get_billing exposes subscription information, plan options, daily consumption history, and spending limits; date parameters let users select a usage-history period. For troubleshooting, scraper_server_status reports instance health, circuit-breaker state, failure counts, and timestamps for successful and failed attempts. The server also retries rate-limited responses according to Retry-After, and access uses an API key tied to a separate Scrapi MCP account.
scrapi-mcp-server setup notes
- scrapi-mcp-server records stdio only; confirm a network transport upstream before planning remote access.