MCP proxy that re-encodes JSON tool responses as GCF. Drop-in, zero changes to your server.

79% fewer tokens than JSON. 90.7% comprehension accuracy where JSON averages 53.6% (1,300+ evals, 10 models, 3 providers). One line change in your MCP config.

pip install gcf-proxy                                         # PyPI
npm install -g @blackwell-systems/gcf-proxy                   # npm
go install github.com/blackwell-systems/gcf-proxy@latest      # Go

gcf-proxy --verbose uvx yfinance-mcp

Use it with any MCP client. When tools return structured JSON, the proxy re-encodes to GCF and logs savings to stderr:

gcf-proxy: get_price_history              54.0KB -> 28.1KB (48% saved)
gcf-proxy: get_ticker_info                10.0KB -> 7.4KB (26% saved)
gcf-proxy: get_price_history              53.8KB -> 27.9KB (48% saved)

--- gcf-proxy session stats ---
Tool calls rewritten:  3
JSON bytes in:         117.8KB
GCF bytes out:         63.4KB
Bytes saved:           54.4KB (46.2%)
Est. tokens saved:     ~13.6K
-------------------------------

Real live stock data from Yahoo Finance. 118KB of JSON reduced to 63KB. ~13,600 tokens saved in 3 tool calls.

Add gcf-proxy in front of any MCP server command in your config:

{
  "mcpServers": {
    "memory": {
      "command": "gcf-proxy",
      "args": ["npx", "-y", "@modelcontextprotocol/server-memory"]
    }
  }
}

Your server keeps outputting JSON. The LLM receives GCF. Nothing else changes.

{"tool":"context_for_task","symbols":[{"qualified_name":"pkg.Auth","kind":"function","score":0.78,...},...]}

GCF tool=context_for_task budget=5000 tokens=1900 symbols=50 edges=20
## targets
@0 fn pkg.Auth 0.78 lsp_resolved
...
## edges [20]
@0<@1 calls

1. Spawns your MCP server as a subprocess
2. Proxies stdin/stdout between client and server
3. Intercepts JSON-RPC responses containing tool results
4. Detects JSON payloads with tool + symbols fields
5. Re-encodes them as GCF
6. Passes everything else through unchanged

Non-convertible responses (plain text, HTML, errors) pass through untouched.

Sometimes you can't. The server is a third-party binary, or it's maintained by another team, or you just don't want to add a dependency. gcf-proxy gives you the token savings without touching server code.

If you control the server, use the GCF libraries directly for better control over session deduplication and delta encoding.

GCF achieves 100% comprehension accuracy at 500 symbols where JSON scores 76.9% and TOON scores 92.3%. On TOON's own benchmark, GCF wins all 6 datasets.

Reproduce comprehension eval: git clone https://github.com/blackwell-systems/gcf-go && cd gcf-go/eval && GOWORK=off go test -run TestComprehension -v -timeout 0

Reproduce token benchmark: git clone https://github.com/blackwell-systems/toon && cd toon && git checkout gcf-comparison && cd benchmarks && pnpm install && pnpm benchmark:tokens

• GCF Specification
• Documentation
• Go library
• TypeScript library
• Python library
• Rust library
• Swift library
• Kotlin library

MIT - Dayna Blackwell / GCF