Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

rtk

rtk• Pinned

rtk: A cost reduction engine for AI coding agents.

What problem it solves

AI coding agents can consume large amounts of input while reading verbose test runs, diffs, logs, and build output. RTK reduces the command output entering an agent's context and provides optional pipelines for additional compression and routing.

Key features

  • Automatic command output compression — RTK rewrites shell commands so common development tools return compact output for agents to read. Failed filters fall back to raw output, and the original exit code is preserved.

    Source: README, src

  • Support for AI coding agents — RTK can integrate through hooks, plugins, or rules files across a range of AI coding tools. Agent setup is performed with `rtk init` variants.

    Source: README, hooks

  • Composable pipelines — Pipelines extend the command compressor with agent-agnostic instructions for transforming what an agent reads and sends. Community pipelines can be installed, while developers can write and share their own.

    Source: README, engine

  • Local savings analytics — RTK tracks commands it touches and reports estimated savings through `rtk gain`. The dashboard can show summaries, history, daily breakdowns, and JSON exports.

    Source: README

  • Optional provider API proxy — The separate `rtk-proxy` daemon can apply a pipeline to provider API traffic while preserving the prompt cache. It also carries calls routed to cheaper models so they appear in the savings dashboard.

    Source: README, proxy

Quick start

Copied verbatim from the project's README.

brew install rtk
rtk init -g                      # Claude Code; every other agent: see "Works in your agent"

Who it may suit

Inference: may suit developers using AI coding agents who want to reduce command-output context without changing their normal shell workflow. It may also suit teams interested in reusable pipelines and local savings reporting.

Topics to discuss

  • Which development commands should be prioritized for additional output filters?

  • Which pipeline transformations are useful across different coding agents?

  • How should estimated savings be measured and interpreted consistently across projects?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/rtk-ai/rtk#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.

github.comhttps://github.com/rtk-ai/rtk

Comments

…