rtk-ai/rtk

📈 Trending★ 49,962RustAudience · developerComplexity · 3/5ActiveLicenseSetup · moderate

Mindmap

mindmap
  root((repo))
    What it does
      Intercepts AI requests
      Compresses context
      Reduces token usage
    How it works
      Sits between tool and LLM
      Hooks into bash layer
      Runs locally
    Use cases
      Lower API costs
      Extend context windows
      Heavy coding sessions
    Tech stack
      Rust
      CLI proxy
    Audience
      Claude Code users
      Cursor users
      Budget-conscious devs

mindmap root((repo)) What it does Intercepts AI requests Compresses context Reduces token usage How it works Sits between tool and LLM Hooks into bash layer Runs locally Use cases Lower API costs Extend context windows Heavy coding sessions Tech stack Rust CLI proxy Audience Claude Code users Cursor users Budget-conscious devs

Things people build with this

USE CASE 1

Reduce token consumption and API costs when using Claude Code, Cursor, or GitHub Copilot for extended coding sessions.

USE CASE 2

Extend effective context window length by compressing requests, allowing longer conversations on large codebases without hitting limits.

USE CASE 3

Optimize AI assistant usage for teams or individuals with strict token budgets or rate-limit constraints.

Tech stack

RustCLIProxy

Getting it running

Difficulty · moderate Time to first run · 30min

Requires Rust toolchain installation and compilation from source.

Use freely for any purpose including commercial. Keep the notice and disclose changes to the patent grant.

In plain English

RTK is a local CLI proxy that sits between your AI coding assistant and the LLM provider it calls, with the goal of dramatically reducing the number of tokens consumed per session. According to the project, it achieves a 60 to 90 percent reduction in token usage, which translates directly into lower API costs or fewer rate-limit hits when working with tools like Claude Code, GitHub Copilot, or Cursor. It works by intercepting the outgoing requests that your AI coding tool makes to its backend model, transforming or compressing the context before forwarding it. Because it hooks into the bash command layer that tools like Claude Code use when invoking shell commands, it can operate without modifying the AI tool itself. The proxy runs locally on your machine and the AI assistant's configuration is updated to route requests through it. You would use RTK when your AI coding sessions are consuming tokens faster than your budget or subscription allows, or when you are hitting context limits that cut off long conversations. It is particularly relevant for heavy users of Claude Code, Codex, or Cursor who run extended sessions on large codebases and find that context window costs or limits become a practical constraint. The project is written in Rust, which gives it low overhead as a local proxy process. It is designed to be transparent to the user: once configured, the AI coding tool behaves the same as before but the underlying requests are more token-efficient. This makes it a cost-optimization layer rather than a change to the AI tool's functionality.

Copy-paste prompts

Prompt 1

How do I set up RTK as a local proxy between my Cursor editor and Claude's API to reduce token usage?

Prompt 2

Show me how to configure my AI coding tool to route requests through RTK without modifying the tool itself.

Prompt 3

What context compression techniques does RTK use to achieve 60, 90% token reduction, and how can I verify the savings?

Open on GitHub → Explain another repo

Generated 2026-05-18 · Model: sonnet-4-6 · Verify against the repo before relying on details.