Love this, brings back LAN party vibes!
Sound notifications for Claude Code are a real pain point.
I built something in the same space but took a different approach — less fun, more engineering:
Vox (https://github.com/rtk-ai/vox) — local TTS in pure Rust, no API key, no cloud dependency.
I use Claude Code all day and got tired of watching it eat tokens on noisy CLI output. git status on a big repo, find across a project, cargo test with 200 lines of compilation — none of that noise is useful to the model.
So I built RTK. It wraps commands and strips the junk before it reaches your context:
rtk git status # instead of git status
rtk find "*.rs" . # compact results
rtk ls . # token-optimized tree
rtk cargo test # just the test results
My rtk gain -q after 15 days of real usage:
Total commands: 7,061
Input tokens: 29.3M
Output tokens: 4.8M
Tokens saved: 24.6M (83.7%)
Biggest surprise was rtk find — directory listings are insanely wasteful. Run rtk discover on your Claude Code session history to see how many tokens you've been wasting — that's what convinced me to actually build this.
Rust, MIT licensed. Happy to hear what commands are the worst token offenders for you.
Hi,
I’m splitting my time between multi-cloud governance and optimizing my "vibe coding" workflow:
Kexa (https://kexa.io)
An open-source multi-cloud governance framework powered by a YAML rule engine. We just reached a milestone by adding Database (DB) support.
The goal is to allow developers to audit configuration and compliance directly within DB instances, alongside standard cloud resources (AWS, GCP, Azure, K8s). We’re focusing on keeping the YAML rules as agnostic as possible so the same logic can apply across different environments without rewriting everything.
rtk (https://github.com/rtk-ai/rtk)
This is a "scratching my own itch" project born from using Claude-code. While vibe coding, I got frustrated watching the agent spam ls -al or cat repeatedly just to "orient" itself.
It creates two main issues:
The Token Tax: It burns through tokens for info the agent already has.
Context Pollution: The context window fills up with redundant noise.
rtk acts as a CLI wrapper/filter to make LLM interactions more signal-to-noise efficient. It silences or summarizes redundant outputs so the agent only receives the necessary "delta." It’s a simple attempt to keep the context clean and make sessions last longer before the agent loses the plot.
v0.7.1 !
all new features:
- Token economics dashboard — track savings per command, cost analysis, execution time
- Session scanner — analyzes your Claude Code history, finds commands that could have used rtk, and tells you how many tokens you wasted
- Auto-rewrite hook — intercepts Claude Code calls in real-time and rewrites them to rtk equivalents, no manual setup needed
- Rust toolchain support — compact output for cargo build, test, clippy
- Git power user commands — branch, fetch, stash, worktree, show, full flag forwarding
- Full JavaScript/TypeScript stack — npm/npx routing, pnpm build/typecheck, curl with auto JSON detection
- GitHub CLI proxy — PR create, merge, diff, comment, edit, gh api passthrough
- Three-tier parser infrastructure — JSON, regex, raw fallback system for reliable output compression across all tools
