Home › Knowledge & Memory › token-compressor

token-compressor

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

Python

★ 12 stars Last pushed 2026-03-08 License: MIT

View repository →

Connect

Review any command before running it. Package names and URLs come from the server's own registry entry.

uvx token-compressor-mcp

Or add to your MCP client config:

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

Up-to-date docs for AI agents

The best-benchmarked open-source AI memory system. And it's free.

Hindsight: long-term memory for AI agents modeled on human memory.

Memory manager for AI apps and agents using graph and vector stores, with ingestion from many data sources.

Turn docs, GitHub repos, PDFs, videos, Confluence, Notion and Slack/Discord into AI skills and RAG knowledge, exportable to vector…

Local-first knowledge management with bi-directional LLM sync via Markdown files.

Data from the Official MCP Registry