Efficient-Token

fahomid/Efficient-Token

★ 1 stars TypeScript AI/LLM Updated today

A local-first MCP server that does deterministic code work on your machine and returns only distilled, faithful results to the model — cutting token usage without degrading reasoning.

View on GitHub → Try with Claude — $10 free →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "efficient-token": {
      "command": "npx",
      "args": [
        "-y",
        "fahomid/Efficient-Token"
      ]
    }
  }
}

README Excerpt

<div align="center"> **A local-first [MCP](https://modelcontextprotocol.io) server that cuts LLM token usage — without degrading reasoning.** It does deterministic code work on your machine and returns only distilled, *faithful* results to the model: real slices, real symbols, a fraction of the tokens.

Tools (20)

EFFICIENT_TOKEN_GROUPSEFFICIENT_TOKEN_MAX_FILE_BYTESEFFICIENT_TOKEN_MAX_READ_TOKENSEFFICIENT_TOKEN_ROOTapply_patchcall_hierarchycall_siteschange_coveragecheck_locatecode_checkcode_contextcode_editcode_outlinecode_readcode_searchcode_writecolor_contrastcommit_logconflict_digestdesign_tokens

Efficient-Token

Quick Install

README Excerpt

Tools (20)

Related AI/LLM Servers