Token Pilot

Name: Token Pilot
Author: Digital-Threads

@Digital-Threads

About Token Pilot

Save 60-80% tokens when AI reads code — MCP server for token-efficient code navigation with AST-aware structural reading

#token-savings #code-reading #ast #mcp #claude-code #cursor #codex #typescript #python #developer-tools

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "token-pilot": {
      "command": "npx",
      "args": [
        "-y",
        "token-pilot"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Token Pilot?

Token Pilot reduces context consumption in AI coding assistants by up to 90% without changing how you work. It provides MCP tools, PreToolUse hooks, and tp-* subagents to replace heavy file reads with structural summaries, symbol lookups, and token-efficient alternatives. Designed for users of Claude Code, Cursor, Codex CLI, Gemini CLI, and Cline.

How to use Token Pilot?

Run npx -y token-pilot init to create or merge the .mcp.json configuration. Restart your AI assistant to activate. For Claude Code, install via the plugin marketplace with claude plugin marketplace add https://github.com/Digital-Threads/token-pilot then claude plugin install token-pilot@token-pilot.

Key features of Token Pilot

23 MCP tools for structural reads, symbol search, git analysis, session analytics
PreToolUse hooks intercept heavy Read, Grep, Bash calls and redirect to efficient alternatives
25 tp-* subagents (Claude Code only) with MCP-first behavior and budget enforcement
Three enforcement modes: advisory, deny (default), strict
Supports TypeScript, JavaScript, Python, Go, Rust, Java, Kotlin, C#, C/C++, PHP, Ruby
Benchmarks show 92% average token savings across four public repos

Use cases of Token Pilot

Load a 15-line outline instead of a 500-line file when you only need structure
Find a specific function by name (read_symbol) without reading the whole module
Block unbounded Grep calls and rewrite them as token-efficient searches
Analyze git history with smart_log capped to 20 commits instead of thousands of lines
Delegate debugging or PR review to a tp-* subagent with a strict token budget

FAQ from Token Pilot

What token savings does Token Pilot deliver?

Up to 90% on average. Benchmarks on four open-source repos (token-pilot, express, fastify, flask) showed 92% input token reduction for files ≥50 lines. Full sessions gain additional savings from session cache, read_symbol, and read_for_edit.

Which AI assistants are supported?

Full support (tools + hooks + subagents): Claude Code. Tools and hooks supported on Cursor, Codex CLI, Gemini CLI, Cline (VS Code). Subagents are Claude Code only. Manual config snippets for each client are in the installation guide.

How do I update Token Pilot?

For Claude Code: claude plugin update token-pilot. For other clients: npx -y token-pilot init always pulls the latest; restart your client. If installed globally, run `npm i -g token-pilot

Comments

More Developer Tools MCP servers

JetBrains MCP Proxy Server

JetBrains

A model context protocol server to work with JetBrains IDEs: IntelliJ, PyCharm, WebStorm, etc. Also, works with Android Studio

962

TalkToFigma

sonnylazuardi

TalkToFigma: MCP integration between AI Agent (Cursor, Claude Code, Codex) and Figma, allowing Agentic AI to communicate with Figma for reading designs and modifying them programmatically.

6.9K