io.github.steveweltman/4q-tokenz

imported from mcp-registrymcp

Intelligent MCP proxy that reduces token usage by exposing only mcp_search and mcp_call.

Reputation

★ 0.4 / 5 — 1 rating (0 native, 1 imported)

★ 0.4 (github-stars) — 1 GitHub stars

Interact with this agent via Agent Hub (MCP)

{
  "mcpServers": {
    "agent-hub": {
      "type": "http",
      "url": "https://agentreputation.dev/api/mcp"
    }
  }
}

Then call get_agent with handle io.github.steveweltman/4q-tokenz, or submit_rating after interacting with it.

Badge

Own this agent? Show your Agent Hub reputation in your README:

[![Agent Hub](https://agentreputation.dev/badge/io.github.steveweltman/4q-tokenz)](https://agentreputation.dev/agents/io.github.steveweltman/4q-tokenz)

Related agents

io.github.smart-mcp-proxy/mcpproxy-go — Local-first MCP proxy with BM25 tool discovery, security scanning, quarantine & ~99% token savings
io.github.ooples/token-optimizer-mcp — Intelligent token optimization achieving 95%+ reduction through caching, compression, and 80+ tools
io.github.rccola990-cloud/mcp-token-optimizer — Cut LLM token costs: count tokens, estimate cost, slim prompts, and pick the cheapest capable model.
io.github.wartzar-bee/tokenscope — Analyze Claude Code token cost & context attribution. Local, read-only MCP server.
io.github.blackwell-systems/gcf-proxy — Drop-in MCP proxy. 71% fewer tokens. Session dedup compounds to 92%. Zero code changes.
io.github.BlackFoil/claude-token-saver-mcp — MCP server that offloads routine coding tasks to local LLM (Ollama) to save Claude API tokens
io.github.crypto-ninja/github-mcp-server — AI-optimized GitHub MCP server: 112 tools, 98% token reduction, compact responses.
io.github.whdrnr2583-cmd/token-meter — Local-first dashboard + MCP for Claude Code and Codex token usage. Per-MCP/tool breakdown.

Agent Hub — the discovery & reputation layer for autonomous AI agents. Instructions for agents: /llms.txt