mcp-link vs fetch
Side-by-side comparison to help you pick between these two MCP servers.
mcp-link by automation-ai-labs | fetch by modelcontextprotocol | |
|---|---|---|
| Stars | ★ 605 | ★ 85,748 |
| 30d uses | — | — |
| Score | 45 | 76 |
| Official | — | ✓ |
| Categories | Developer ToolsWeb ScrapingOther | Web ScrapingAI / LLM ToolsProductivity |
| Language | Go | TypeScript |
| Last commit | 14 mo ago | this month |
mcp-link · Summary
MCP Link automatically converts any OpenAPI V3 API into a fully-functional MCP server.
fetch · Summary
An MCP server that fetches web content and converts HTML to markdown, allowing LLMs to read web pages.
mcp-link · Use cases
- Convert existing REST APIs to MCP servers for AI agent integration without code changes
- Rapidly prototype MCP interfaces for new APIs using OpenAPI specifications
- Unify API access across different services through a standardized MCP interface
fetch · Use cases
- LLMs reading news articles and blogs
- Content analysis of web pages
- Retrieving information from public websites
- Chunked reading of large web documents
mcp-link · Install
Installation
# Clone repository
git clone https://github.com/automation-ai-labs/mcp-link.git
cd mcp-link
# Install dependencies
go mod downloadRunning
# Specify port
go run main.go serve --port 8080 --host 0.0.0.0Usage in Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"openapi-mcp": {
"url": "http://localhost:8080/sse?s=[OpenAPI-Spec-URL]&u=[API-Base-URL]&h=[Auth-Header]:[Value-Prefix]"
}
}
}fetch · Install
Installation
**Using uv (recommended)** No specific installation needed. Use uvx to run the server directly:
uvx mcp-server-fetch**Using PIP** Install via pip:
pip install mcp-server-fetchThen run as:
python -m mcp_server_fetchClaude Desktop Configuration
{
"mcpServers": {
"fetch": {
"command": "uvx",
"args": ["mcp-server-fetch"]
}
}
}