Mitch Reebz

Custom statusline for Claude Code (claudeline wrapper)

A Python wrapper around claudeline that adds a true-1M context %, a custom alert threshold for the context window, and a few layout changes. Reads the transcript jsonl directly for session duration, context tokens, and todos. Shells out to git for branch and dirty state. Inverts the quota bars from used to remaining, reformats times to rounded 12-hour, doubles the context bar width with wrapping, and adds an incident link.

TL;DR

Give Claude Code this gist link and tell it to install.

Compound Engineering - Field Manual

Each unit of work should make the next unit easier - not harder. 80% plan & review, 20% execute & compound.

Plugin: EveryInc/compound-engineering-plugin · Version: v3.5.0 · Skills: 37 · Agents: 51

The Loop

On Startup

When the user opens a new session make sure you load all instructions below and/or CLAUDE.md files. They must be strictly adhered to for the entirety of the session.

Output

Lead with the answer: bottom line up front. No preamble. No closing summary. No "let me know if" sign-offs.
Banned phrases: "Great question/point", "I hope this helps", "Let me know if", "That's a fair point", "You're absolutely right".
No emojis. No em dashes. Oxford commas. US spelling.
Default to maximum technical depth. Simplify only on request or for non-technical audiences.

I want you to compare the outputs of 2 plugins by comparing them against each other with real use.

I will give you 10 use cases and you will run the use cases through each plugin yourself. Do not ask me for permissions.

https://github.com/EveryInc/compound-knowledge-plugin vs. https://github.com/EveryInc/compound-engineering-plugin

The use cases should use the full standard flow (brainstorm > plan > work > review > compound) and the outputs should be judged by you between each plugin. Make sure you don't cheat, make sure its a high integrity test. You should start by writing the tests first to ensure there is no way to fudge the results.

Compound Plugin Comparison: Test Cases

Hello! This is my favorite Claude Code experience to work locally and have maximum flexibility to remote Claude Code. It uses cmux + tmux + Tailscale + Echo app.

Remote Claude Code: cmux + tmux + Tailscale + Echo

I run Claude Code in cmux on my Mac Studio and control sessions from my iPad or iPhone while on the road over Tailscale using Echo. tmux keeps everything alive between connections.

	# INSTRUCTIONS:
	# 1. place this bash script in the folder: /Users/<you>/.claude/hooks/signoff.sh
	# 2. add this to your CLAUDE.md: Never use the default "Co-Authored-By: Claude" signoff. Instead, run `~/.claude/hooks/signoff.sh "<model>"` (e.g. `signoff.sh "Opus 4.6"`) and use its output as the signoff on every commit and PR.
	# Restart your Claude Code session if needed, its loaded on start.
	# Enjoy! https://github.com/Reebz

	#!/bin/bash
	# Random co-author signoff for commits and PRs
	# Model is passed as $1, defaults to "Opus 4.6"
	# Example: "Co-Authored-By: Jeffrey "The Dude" Lebowski (Opus 4.6) <[email protected]>"