Smithery Ai Fetch
Let Claude Code or Cursor agents fetch live webpage HTML and text without custom scrapers during research or build tasks.
Overview
ai.smithery/smithery-ai-fetch is a MCP server for the Build phase that performs agent-driven HTTP fetch requests to webpages via Smithery’s hosted streamable-HTTP endpoint.
What is this MCP server?
- Streamable-HTTP MCP remote at server.smithery.ai with Smithery Bearer auth
- Performs fetch requests to arbitrary webpages for agent tool calls
- Smithery-hosted—no local Node process to maintain
- v1.0.0 server schema from smithery-ai/mcp-servers fetch subfolder
- Single focused tool surface for read-only page retrieval
- Server version 1.0.0 per MCP server.schema.json metadata
- One streamable-http remote endpoint on server.smithery.ai
- Required secret header: Authorization Bearer {smithery_api_key}
What problem does it solve?
Agents cannot reliably read arbitrary live URLs unless you bolt on custom fetch code or overload browser automation for simple GETs.
Who is it for?
Indie builders who want one-line MCP registration for occasional page reads in Claude Code or Cursor without self-hosting.
Skip if: Teams that need JavaScript rendering, authenticated sessions, rate-limited crawl pipelines, or air-gapped local-only MCP.
What do I get? / Deliverables
After you register the remote with your Smithery API key, your agent can request webpage content through MCP tools during build and research tasks.
- Registered remote MCP connection to @smithery-ai/fetch
- Agent-callable fetch tool against arbitrary webpage URLs
- Live HTTP response content returned into the agent context
Recommended MCP Servers
Journey fit
Fetch sits in Build because solo builders wire it into agent workflows while implementing features that need on-demand web reads. Integrations is the right shelf: this is a hosted MCP bridge to HTTP, not frontend UI or shipping checks.
How it compares
Hosted MCP fetch tool, not a full Playwright-style browser skill or scraping framework.
Common Questions / FAQ
Who is ai.smithery/smithery-ai-fetch for?
Solo and indie builders using AI coding agents who need simple webpage retrieval during integrations, validation, or research without maintaining fetch infrastructure.
When should I use ai.smithery/smithery-ai-fetch?
Use it when an agent must read a public URL’s response during build or idea-phase competitor checks and a dedicated browser MCP is overkill.
How do I add ai.smithery/smithery-ai-fetch to my agent?
Add the Smithery remote URL https://server.smithery.ai/@smithery-ai/fetch/mcp as streamable-http MCP and set Authorization to Bearer your Smithery API key.