Token Compressor MCP Server: Install & Setup

Install Token Compressor in Claude Code

Claude Code

claude mcp add token-compressor -- uvx token-compressor-mcp

mcp.json (Claude Desktop, Cursor, VS Code, Windsurf)

{
  "mcpServers": {
    "token-compressor": {
      "command": "uvx",
      "args": [
        "token-compressor-mcp"
      ]
    }
  }
}

Compress prompts 40-60% using local LLM + embedding validation. Preserves all conditionals.

Yes, most likely. Open source and no paid account or API key is mentioned in the listing. Check the provider for current pricing.

Local. Runs on your machine (stdio). You install it with npx, uvx, Docker or from source.

Run: claude mcp add token-compressor -- uvx token-compressor-mcp

LuiggiVal08/toon-memory

Offline persistent memory for AI coding agents — 35 tools, 80% fewer calls, no LLM

var-gg/pindoc

Code-pinned team memory for AI coding agents — typed artifacts, MCP-native, self-host.

juliangeymonat-jpg/wikimoth

Deterministic, token-minimal agent memory: multi-hop [[wikilink]] recall with no LLM in the loop.

arjunkumargiri/cortex

Shared, persistent memory for AI agents — works with Claude, Kiro, Cursor, or any MCP client.

xkumakichi/xaip-protocol

AI agents get on-chain identity, credentials, reputation, escrow, and persistent memory on XRPL.

bluecolumnconsulting-lgtm/bluecolumn-mcp

Persistent memory for AI agents — audio intelligence + Music Memory ingestion and recall