prism

🥈Silver

Prism is a fast, customizable status line for Claude Code. It provides real-time updates and visual feedback for operations teams. Integrates with Claude Code plugins to enhance workflow visibility.

3120Updated 2mo ago

Intermediate30min to implementautomation

Saves ~15 min per use

Quick InstallView Source

git clone https://github.com/himattm/prism.git

Works with:

Claude

Overview

About This Skill

Prism is a native Go status line for Claude Code that displays actionable real-time context usage, git branch information, model details, and device status. It executes plugins in parallel for performance and supports a three-tier configuration system for personal, team, and global settings. Prism automatically detects your billing type to show either API costs with cache efficiency and burn rate, or Max/Pro plan usage limits with countdown timers. The tool includes built-in plugins for Android devices, Spotify, agent task queues, and automatic updates, while also supporting custom plugins written in any language. Operations teams and developers benefit from enhanced workflow visibility and context awareness directly in their Claude Code interface.

How to Use

1. **Install Prism**: Add the Prism plugin to your Claude Code environment using `claude plugins install prism`. Ensure Claude Code is updated to the latest version. 2. **Configure Metrics**: Define the metrics you want to monitor in your `prism.config.json` file. Include thresholds for warnings and critical alerts (e.g., `{ "input_records": { "threshold_warning": 90, "threshold_critical": 95 } }`). 3. **Run with Visibility**: Execute your operation with Prism enabled. Use `claude run --with-prism [SCRIPT]` or integrate Prism into your existing Claude Code workflows. 4. **Monitor in Real-Time**: Watch the status line update as your operation progresses. Prism will highlight anomalies in color (e.g., red for critical, yellow for warnings). 5. **Take Action**: Use the suggested actions from Prism to troubleshoot or optimize your workflow. Adjust thresholds or metrics as needed based on observed patterns.

Use Cases

Monitor API billing costs, cache efficiency, and burn rate in real-time while working in Claude Code

Track Max/Pro plan usage limits with countdown timers for session, weekly, and Opus-specific constraints

Display git branch status, uncommitted changes, and Android device information in the status line

Configure team-wide status line settings via committed repo config while allowing personal overrides

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/himattm/prism

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Use Prism to monitor and display real-time status for [OPERATION]. Configure the status line to show [METRICS] such as [EXAMPLE_METRIC_1], [EXAMPLE_METRIC_2], and [EXAMPLE_METRIC_3]. Highlight any anomalies or thresholds breached in [COLOR]. Provide a summary of the current state and suggest corrective actions if needed.

Example Output

```
🔄 [OPERATION: Data Pipeline Sync] | Status: RUNNING | Duration: 42m 15s
📊 Metrics:
  • Input Records: 12,487/15,000 (83%) [⚠️ WARNING: 90% threshold]
  • Output Records: 11,234/15,000 (75%) [❌ CRITICAL: Lag detected]
  • Errors: 42 (0.34%) [✅ OK: Below 1% threshold]
  • Processing Speed: 342 records/min [⚡ ACCELERATING]
🚨 Alerts:
  - Output lag detected (12m behind schedule)
  - Input records nearing capacity (1,513 remaining)
💡 Suggested Actions:
  1. Pause non-critical jobs to free resources
  2. Investigate output bottleneck in 'transform' stage
  3. Prepare scale-up for input buffer (current: 80%)
```

Current Status: The data pipeline sync is actively running but showing critical output lag. The system has processed 83% of input records but only 75% of output records, indicating a potential bottleneck in the transformation phase. The error rate remains acceptable, but the lag suggests resource contention. The processing speed is improving, which may help close the gap. Immediate intervention is needed to prevent data loss or downstream failures.