kehvinbehvin/json-mcp-filter
by Various
JSON MCP server to filter only relevant data for your LLM
MCP
kehvinbehvin/json-mcp-filter
Added 1 June 2026
Overview
An open source MCP server written in JavaScript that filters JSON data to extract only relevant fields for an LLM. It helps reduce token usage and noise by allowing developers to specify which parts of a JSON structure are passed to the language model.
Best for
Best for
Developers using MCP to optimize JSON data sent to LLMs, especially in prototyping or low‑traffic projects
Use cases
- Stripping irrelevant keys from large JSON payloads before sending to an LLM
- Reducing token consumption when feeding structured data into a context window
- Enforcing a whitelist of fields for consistent LLM input across endpoints
Notes
An open source MCP server written in JavaScript that filters JSON data to extract only relevant fields for an LLM. It helps reduce token usage and noise by allowing developers to specify which parts of a JSON structure are passed to the language model.
24 stars on GitHub. Last updated 2026-03-30. Licensed MIT.
Use cases
- Stripping irrelevant keys from large JSON payloads before sending to an LLM
- Reducing token consumption when feeding structured data into a context window
- Enforcing a whitelist of fields for consistent LLM input across endpoints
Pros
- Lightweight and straightforward JavaScript implementation
- Open source with a permissive license for modification
- Targets a specific pain point in MCP workflows
Cons
- Small community (24 stars) may mean limited support or slower updates
- Requires familiarity with Model Context Protocol ecosystem
- No built-in field validation or error handling beyond basic filtering
Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.
Pros
- Lightweight and straightforward JavaScript implementation
- Open source with a permissive license for modification
- Targets a specific pain point in MCP workflows
Cons
- Small community (24 stars) may mean limited support or slower updates
- Requires familiarity with Model Context Protocol ecosystem
- No built-in field validation or error handling beyond basic filtering
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
agent-infra/mcp-server-browser
Various
The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra
vectorize-io/hindsight
Various
Hindsight: Agent Memory That Learns
microsoft/markitdown
Various
Python tool for converting files and office documents to Markdown.