A colorful, informative statusline for Claude Code CLI.
- Directory (blue) - Current working directory
- Node version (green ⬢) - Shows when package.json exists
| #!/usr/bin/env bash | |
| set -euo pipefail | |
| # patch-claude-code.sh — Rebalance Claude Code prompts to fix corner-cutting behavior | |
| # | |
| # What this does: | |
| # Patches the npm-installed @anthropic-ai/claude-code cli.js to rebalance | |
| # system prompt instructions that cause the model to cut corners, simplify | |
| # excessively, and defer complicated work. | |
| # |
| // Add this to `.claude/settings.json` to replace Claude Code's silly spinner verbs with something a bit harder-hitting. | |
| "spinnerVerbs": {"mode": "replace", "verbs": [ | |
| "The electric fence stopped working years ago", | |
| "You already know the answer", | |
| "Comfort is the expensive kind of stuck", | |
| "The map is not the territory", | |
| "You're rehearsing a fear, not a fact", | |
| "What you resist persists", | |
| "The obstacle is the way", |
When analyzing large codebases or multiple files that might exceed context limits, use the Gemini CLI with its massive
context window. Use gemini -p to leverage Google Gemini's large context capacity.
Use the @ syntax to include files and directories in your Gemini prompts. The paths should be relative to WHERE you run the
gemini command:
| Model | Quota (Plus Subscription) | Typical Use Cases |
|---|---|---|
| GPT-4o | 80 messages per 3 hours ([help][1]) | Multimodal, high-quality conversational AI |
| o4-mini | 300 messages per day ([help][2]) | Coding, math/science problems, tool-based tasks |
| o4-mini-high | 100 messages per day ([help][2]) | Frequent, moderate-complexity tool-assisted work |
| o3 | 200 messages per week ([help][2]) | General-purpose reason |
| import React from 'react'; | |
| const MIN_SCALE = 1; | |
| const MAX_SCALE = 4; | |
| const SETTLE_RANGE = 0.001; | |
| const ADDITIONAL_LIMIT = 0.2; | |
| const DOUBLE_TAP_THRESHOLD = 300; | |
| const ANIMATION_SPEED = 0.04; | |
| const RESET_ANIMATION_SPEED = 0.08; | |
| const INITIAL_X = 0; |
| [ -s $HOME/.nvm/nvm.sh ] && . $HOME/.nvm/nvm.sh # This loads NVM | |
| function node { | |
| rawnode=$(which node) | |
| icu4cdata=$(dirname $(dirname $rawnode))/lib/node_modules/icu4c-data | |
| [[ -d $icu4cdata ]] || npm install -g icu4c-data@$($rawnode -e ' | |
| console.log((v => v.icu_ver_major + v.icu_endianness)( | |
| process.config.variables));') | |
| NODE_ICU_DATA=$icu4cdata $rawnode "$@" | |
| } |
| import React, { Component } from 'react' | |
| import Subapp from './subapp/Root' | |
| class BigApp extends Component { | |
| render() { | |
| return ( | |
| <div> | |
| <Subapp /> | |
| <Subapp /> | |
| <Subapp /> |
A small function in response to Taylor Hunt's blog post "Optimizing svgs in data uris"* allowing you to skip base64 encoding and simply paste the SVG markup right in the CSS.
Forked from jakob-e's Pen Encode SVG SCSS.
A Pen by Max Dolgov on CodePen.
| var pureRender = (Component) => { | |
| Object.assign(Component.prototype, { | |
| shouldComponentUpdate (nextProps, nextState) { | |
| return !shallowEqual(this.props, nextProps) || | |
| !shallowEqual(this.state, nextState); | |
| } | |
| }); | |
| }; | |
| module.exports = pureRender; |