How does this reduce token usage?
By extracting only needed data using JSONPath patterns instead of fetching entire JSON responses, saving up to 99% of tokens.
@ackness
Basic information
A Model Context Protocol (MCP) server that provides tools for fetching and extracting JSON data from URLs using JSONPath patterns.
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"fetch-jsonpath-mcp": {
"command": "uvx",
"args": [
"fetch-jsonpath-mcp"
]
}
}
}Extract JSON content from a URL using JSONPath with extended features. Supports extensions like len, keys, filtering, arithmetic operations, and more. If 'pattern' is omitted or empty, the entire JSON document is returned. Supports different HTTP methods (default: GET).
Fetch text content from a URL using various HTTP methods. Defaults to converting HTML to Markdown format.
Batch extract JSON content from multiple URLs with different extended JSONPath patterns. Supports all JSONPath extensions and optimizes by fetching each unique request only once. Executes requests concurrently for better performance. Supports different HTTP methods.
Batch fetch raw text content from multiple URLs using various HTTP methods. Executes requests concurrently for better performance.
Fetch Jsonpath Mcp is a Model Context Protocol (MCP) server that provides tools for fetching JSON data and web content from URLs. It features intelligent content extraction, multiple HTTP methods, and browser-like headers for reliable web scraping, designed to reduce LLM token usage and hallucination by extracting only needed data via JSONPath patterns.
Install in any MCP-compatible IDE by adding the server configuration with command uvx and args ["fetch-jsonpath-mcp"]. For Claude Code, run claude mcp add fetch-jsonpath-mcp -- uvx fetch-jsonpath-mcp. Tools include fetch-json, fetch-text, batch-fetch-json, and batch-fetch-text. Customize behavior via environment variables such as JSONRPC_MCP_TIMEOUT, JSONRPC_MCP_VERIFY, JSONRPC_MCP_HEADERS, and JSONRPC_MCP_PROXY.
By extracting only needed data using JSONPath patterns instead of fetching entire JSON responses, saving up to 99% of tokens.
All common HTTP methods: GET, POST, PUT, DELETE, PATCH, HEAD, and OPTIONS.
Set environment variables `JSONRPC_MCP_HEADERS` for custom headers (merged with default browser headers) and `JSONRPC_MCP_PROXY` for an HTTP proxy.
Three formats: `markdown` (default), `clean_text` (pure text with HTML tags removed), and `raw_html` (original HTML).
Yes, use the `batch-fetch-json` and `batch-fetch-text` tools to process multiple URLs concurrently.
๐ฅ Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients.
A flexible HTTP fetching Model Context Protocol server.

Parse allows your agent to access thousands of APIs instantly. You can use Parse to build an API for any action on any website and edit any previously existing API in our catalogue.
๐ฎ Platform-neutral Browser Use for AI agents: real Chrome automation with a CLI + SDKs, no lock-in, dead simple.
Capture screenshots of webpages as images or PDFs with Screenshot Scout.
Comments