token-compressor
Compress prompts 40-60% using local LLM + embedding validation. Preserves all conditionals.
stdio v0.1.1
Connection Configuration
Add this to your AI agent's MCP config file to connect.
Run in your terminal:
claude mcp add io.github.base76-research-lab--token-compressor -- uvx
Server Details
Transport
stdio
Authentication
NoneVersion
v0.1.1
Server Name
io.github.base76-research-lab/token-compressor
Last Updated
Mar 3, 2026
Get Started
How to install and connect this MCP server.