buildatscale-claude-code

BuildAtScale Claude Code Plugins

Claude Code plugins from BuildAtScale - slash commands, hooks, and skills for enhanced productivity.

Quick Start

Add the Marketplace

/plugin marketplace add https://github.com/buildatscale-tv/claude-code-plugins

Install Plugins

Core features (slash commands and hooks):

/plugin install buildatscale@buildatscale-claude-code

Image generation skill:

/plugin install nano-banana@buildatscale-claude-code

Promo video creation skill:

/plugin install promo-video@buildatscale-claude-code

Available Plugins

buildatscale (Core Tools)

/plugin marketplace add https://github.com/buildatscale-tv/claude-code-plugins
/plugin install buildatscale@buildatscale-claude-code

Core slash commands and hooks for git workflow automation.

Slash Commands:

• /buildatscale:commit - Create commit message(s) for staged/unstaged changes, breaking into logical units
• /buildatscale:pr - Create pull request with GitHub CLI, auto-branching from main/master
• /buildatscale:ceo - Create executive summary of work in progress, recent work, or recently deployed changes

Hooks:

• bash-guard.sh - Blocks dangerous bash commands (sudo, credential access, disk ops, exfiltration, etc.)
• file-guard.sh - Blocks writes to system directories, config files, and credential files
• file-write-cleanup.sh - Cleans up files after write/edit operations
• git-block-force-push.sh - Prevents dangerous git operations like force push

Scripts:

• statusline.sh - Enhanced status line with context runway gauge (see below)

Statusline Setup

The statusline script provides an enhanced status display with:

• Context runway gauge - Shows remaining context (not used), so you know how much runway you have left
• Color-coded warnings - Optional coloring: Green (OK) → Yellow (low) → Red (critical, compaction needed)
• Configurable thresholds - Set custom warning/critical percentages
• Git branch display - Current branch in green
• Relative path display - Shows ./project/subdir when in subdirectories
• Configurable cost display - Toggle on/off for API users

Example output:

[./website/src][main]              +156/-23 | ████████████░░░░░░░░ 58% | Opus

To enable, add to your ~/.claude/settings.json:

{
  "statusLine": {
    "type": "command",
    "command": "bash ~/.claude/plugins/marketplaces/buildatscale-claude-code/plugins/buildatscale/scripts/statusline.sh"
  }
}

Note: Status line is configured as a top-level setting, not through the plugin system. This must be added manually after installing the plugin.

Flags (append to the command string):

Flag	Description	Default
--display <free|used>	What to show: free (runway left) or used (consumed)	free
--detail <full|minimal>	full (progress bar + %) or minimal (just %)	minimal
--color-usage	Colorize context gauge (green/yellow/red)	off (gray)
--color-usage warnings	Only color at warning/critical levels (gray when OK)	—
--usage-warning <pct>	Free % threshold for warning color	25
--usage-critical <pct>	Free % threshold for critical color	10
--no-color	Disable all ANSI colors/formatting	off
--cost	Show session cost (useful for API users)	off

Example with flags:

{
  "statusLine": {
    "type": "command",
    "command": "bash ~/.claude/plugins/marketplaces/buildatscale-claude-code/plugins/buildatscale/scripts/statusline.sh --detail full --color-usage warnings --cost"
  }
}

nano-banana (Skill)

/plugin marketplace add https://github.com/buildatscale-tv/claude-code-plugins
/plugin install nano-banana@buildatscale-claude-code

Generate images using Google's Gemini models (Nano Banana). See the demo video.

Available Models:

Model	Flag	Best For	Max Resolution
Flash (Nano Banana)	flash	Speed, high-volume tasks	1024px
Pro (Nano Banana Pro)	pro	Professional quality, complex scenes	Up to 4K
2 (Nano Banana 2)	2	Fast + high-res, best all-around	Up to 4K

Prerequisites:

• uv - Python package manager (required to run the image generation script). See the uv installation walkthrough
• GEMINI_API_KEY environment variable with your Google AI API key

Usage:

uv run "${SKILL_DIR}/scripts/image.py" \
  --prompt "Your image description" \
  --output "/path/to/output.png"