Curated GitHub repositories, open-source tools, and 566 brand DESIGN.md specifications for AI agents.
Real-time Claude Code context usage monitor with visual progress bars, color-coded alerts, session analytics (cost, duration, lines changed), and auto-compact warnings. Tracks conversation context consumption and provides visual feedback to prevent session interruptions.
Connect Claude Code with Google Vertex AI to access Anthropic's Claude models through Google Cloud Platform. Automatically configures all available Claude models (Sonnet, Haiku, Opus) with enterprise-grade infrastructure, billing, and security. Requires: GCP project with Vertex AI API enabled, authenticated gcloud CLI, and model access approval in Model Garden.
Real-time Vercel deployment monitor with clickable deploy link. Shows build status, time since last deployment, and a clickable OSC 8 hyperlink to the deployment URL (Cmd click). Setup: Export environment variables 'export VERCEL TOKEN your token' and 'export VERCEL PROJECT ID your project id' (or manually replace $VERCEL TOKEN and $VERCEL PROJECT ID in the command if you prefer not to use environment variables). Get your token from vercel.com account tokens and project ID from your Vercel dashboard.
Sonnet as the main model, Opus as the advisor it consults at decision points, and Haiku for every subagent, including agents whose definition pins another model (remove CLAUDE CODE SUBAGENT MODEL FORCE to let them keep theirs). Uses aliases, so it follows the latest version of each family. The advisor needs the Anthropic API and stays off when DISABLE TELEMETRY or CLAUDE CODE DISABLE NONESSENTIAL TRAFFIC is set; the force flag needs Claude Code v2.1.257.
Configure a custom status line using a shell command that receives session context via JSON stdin. The script can display model name, current directory, git branch, or any dynamic information. Create your script at .claude statusline.sh and make it executable.