Agent♥︎Age
Catalog

web-retrieval-mcp

Official

by VelvetSP · Python

MCP web search + tiered web fetch for AI agents (Exa, Firecrawl), SSRF-guarded, cross-platform.

web-retrieval-mcp (io.github.VelvetSP/web-retrieval-mcp)

The io.github.VelvetSP/web-retrieval-mcp server is an open-source Model Context Protocol (MCP) server for source-grounded web research. It provides MCP access to web search and tiered web fetching for AI agents, including Exa and Firecrawl workflows, with SSRF guarding and cross-platform support. It focuses on read-only retrieval and provenance.

🛠️ Key Features

  • Model Context Protocol (MCP) server
  • Six read-only tools for web search and web fetch
  • Exa search support and resilient retrieval chain (Exa → Camoufox → Tavily → Firecrawl)
  • AI/ML paper discovery and developer-source search
  • Explicit provenance and optional local caching
  • SSRF-guarded design
  • Cross-platform operation

🚀 Use Cases

  • Source-grounded web research for AI agents
  • Connecting MCP clients (e.g., Codex CLI/IDE extension, Claude Code/Desktop, Cursor, ChatGPT desktop)
  • Discovering AI/ML papers and developer resources via search and page retrieval

⚡ Developer Benefits

  • Integrates with compatible MCP clients
  • Resilient page fetching across multiple providers
  • Read-only tool access with explicit provenance
  • Optional local cache support for retrieval resilience

⚠️ Limitations

  • The server exposes read-only tools only
  • Provider support is described as centered on Exa and Firecrawl (and the cited retrieval chain)

Topics

ai-agentscamoufoxclaude-codeexafirecrawlllm-toolsmcpmcp-servermodel-context-protocolsearch-apiweb-fetchweb-scrapingweb-searchcodexcodex-clicursorcursor-aicursor-plugingrokgrok-cli

Related servers

More in Search & Web