#18 · Primary category: AI Coding Assistants
rtk
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
Project last updated:08/29/26
GitHub Stars
77.8K
Forks
4.9K
Contributors
135
License
Apache-2.0
Why we included this project
Anyone who runs an agentic coding tool has watched the context window fill with raw shell output: full git diffs, verbose test logs, and one-line-per-file listings. RTK sits between the agent and the shell, intercepting commands and compressing their output before the LLM reads it. It understands over a hundred common commands, so `git status` collapses into a grouped stat line, `cargo test` reports only failures, and `ls` becomes a tree with file counts. That cuts input tokens per command by 60 to 90 percent, which helps long sessions stay inside context limits and keeps API costs down. It ships as a single Rust binary with no dependencies and hooks into most major coding agents automatically, so it slots into an existing AI-assisted workflow without much setup.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
superpowers
An agentic skills framework & software development methodology that works.
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
skills
Skills for Real Engineers. Straight from my .agents directory.
opencode
The open source coding agent.
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.