Refetch provides web content retrieval and search capabilities optimized for AI agents. The service excels at converting messy web content into clean, AI-readable formats while offering both web and news search functionality.
Core Capabilities
Content Extraction: Fetch any webpage and receive clean markdown output, perfect for analysis and summarization. Can also retrieve raw HTML when you need to inspect source code or extract specific elements, plus structured JSON-LD data for machine-readable schema information.
Search Operations: Perform thorough web searches to find information and answer questions. Dedicated news search focuses on recent articles and breaking news coverage.
Usage Monitoring: Check your current credit balance and API usage history without consuming credits.
Content Formats
The service returns web content in three formats depending on your needs:
- Markdown: Clean, readable text ideal for summarization and analysis
- HTML: Raw source for technical inspection or specific element extraction
- JSON-LD: Structured schema.org data for products, articles, events, and other entities
Available Templates
The server provides prompt templates for common workflows: finding latest news coverage on topics, researching subjects with structured overviews, and summarizing webpage content.
Usage Notes
Credit-based pricing model — monitor your balance regularly. The markdown extraction is the primary tool for most content needs. Use HTML extraction only when you specifically need to inspect source code or page structure.
