maxylev/searchfetch

Overview

A fault-tolerant MCP server for stealth web search and fetching. It provides an interface for AI agents to search the web and retrieve pages while avoiding detection. Built in Python, it integrates with clients that support the Model Context Protocol.

Best for

Best for
Developers building MCP-compatible AI agents that need reliable, low-profile web access

Use cases

• Add web search to MCP-based AI agents
• Fetch web pages without triggering bot detection
• Build resilient search tools that handle failures gracefully

How to use

Tools exposed

• max_results
• safe_search
• block_media
• start_index
• max_length

Tested with

Claude Desktop, Claude Code, Cursor

Example client config

{\n  "mcpServers": {\n    "searchfetch": {\n      "command": "uvx",\n      "args": ["searchfetch"]\n    }\n  }\n}

Notes

A fault-tolerant MCP server for stealth web search and fetching. It provides an interface for AI agents to search the web and retrieve pages while avoiding detection. Built in Python, it integrates with clients that support the Model Context Protocol.

3 stars on GitHub. Last updated 2026-09-15. Licensed MIT.

Use cases

• Add web search to MCP-based AI agents
• Fetch web pages without triggering bot detection
• Build resilient search tools that handle failures gracefully

Pros

• Fault-tolerant design handles search and fetch failures
• Stealth features reduce risk of being blocked
• Simple Python implementation for MCP integration

Cons

• Very low adoption with only 3 stars
• Limited documentation and community support
• Stealth capabilities may raise ethical or policy concerns

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.