Token Saver

An MCP server plugin for Claude Code that automatically reduces token usage across your sessions. All optimizations are purely algorithmic — no extra API calls, no added cost.

Tools

All tools return plain text with a compact stats footer — results are deliberately not JSON-wrapped, since JSON escaping of newlines and quotes would inflate the very token count this server exists to reduce.

Note on token counts: the server uses the cl100k_base encoding (via tiktoken), which is OpenAI's tokenizer. Claude's tokenizer differs, so all counts are approximations — typically within ~10–20% of Claude's actual usage. Relative savings percentages are unaffected.

Benchmark results

Measured against real code fixtures and realistic prompt inputs. See benchmark/BENCHMARK.md for full methodology.

* the smart_read_file average includes tiny synthetic fixtures used as multi-language correctness tests; on realistic files it ranges 38–81%.

Run the benchmark yourself:

npm run benchmark

Installation

1. Clone and install

git clone <your-repo-url> token-saver-mcp
cd token-saver-mcp
npm install

2. Add to Claude Code

claude mcp add token-saver -- node /absolute/path/to/token-saver-mcp/src/index.js

Or manually edit ~/.claude.json under mcpServers:

{
  "mcpServers": {
    "token-saver": {
      "command": "node",
      "args": ["/absolute/path/to/token-saver-mcp/src/index.js"]
    }
  }
}

3. Verify

claude mcp list

You should see token-saver listed as connected.

Usage examples

Use smart_read_file on src/api/routes.js, focus on "authentication" and "middleware"

Generate a .claudeignore for my project at /home/user/myapp and write it to disk

Count tokens in this output: [paste output]

Compress this before sending: [paste code]