ccs

🥈Silver

CCS enables switching between Claude, Gemini, Copilot, and 300+ models via CLI, ProxyAPI, or OAuth. It benefits developers and operations teams by providing a visual dashboard, remote proxy support, and WebSearch fallback. The tool connects to AI models and workflows, streamlining multi-account management and reducing setup time.

1,912810Updated 3w ago

Intermediate30min to implementautomation

Saves ~240 min per use

Quick InstallView Source

git clone https://github.com/kaitranntt/ccs.git

Works with:

Claude

Overview

About This Skill

https://docs.ccs.kaitran.ca

How to Use

[{"step":"Install and configure CCS by following the official setup guide for your environment (CLI, ProxyAPI, or OAuth).","tip":"Use `ccs --list-models` to verify available models before switching."},{"step":"Switch to your desired model using the CCS CLI or API. For CLI: `ccs switch --model [MODEL_NAME]`. For API: call the `/switch-model` endpoint with the model parameter.","tip":"Enable WebSearch fallback if the primary model may lack real-time data or domain-specific knowledge."},{"step":"Execute your task or workflow. Use the model's native capabilities (e.g., code execution, tool use) as needed.","tip":"For complex tasks, break them into subtasks and switch models mid-stream if required (e.g., switch to a code-focused model for analysis)."},{"step":"Review the output and iterate. Use CCS's dashboard to monitor model performance, token usage, and fallback events.","tip":"Log fallback events to identify persistent gaps in model capabilities or data freshness."}]

Use Cases

Run simultaneous sessions for work and personal Claude accounts to maintain separation.

Quickly switch between different AI models like Gemini and Copilot for diverse project needs.

Utilize the dashboard to monitor the health and status of all connected AI profiles in real-time.

Implement parallel workflows by executing tasks across multiple terminals with different AI providers.

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/kaitranntt/ccs

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 the CCS tool to switch to [MODEL_NAME] for [TASK_DESCRIPTION]. If the model isn't available, fall back to [FALLBACK_MODEL] with WebSearch enabled. Execute the task and provide the output in [FORMAT]. Example: 'Use CCS to switch to gemini-1.5-pro for analyzing the customer feedback dataset. Fall back to claude-3-7-sonnet if unavailable, with WebSearch enabled. Provide the output as a structured CSV with sentiment scores.'

Example Output

```
Model: gemini-1.5-pro (switched via CCS CLI)
Task: Analyze customer feedback dataset from Q2 2024 (2,458 entries) to identify top pain points and sentiment trends.

Results:
1. **Top Pain Points** (by frequency):
   - Shipping delays (38% of negative feedback)
   - Product quality issues (22%)
   - Pricing concerns (15%)

2. **Sentiment Distribution**:
   - Positive: 42% (↑ 8% from Q1)
   - Neutral: 31%
   - Negative: 27% (↓ 5% from Q1)

3. **Key Insights**:
   - Customers mentioning "shipping" in negative feedback are 3x more likely to churn.
   - Product quality complaints correlate with lower NPS scores (-12 points).

Recommendations:
- Prioritize shipping reliability improvements (e.g., partner with 3PL for faster delivery).
- Launch a quality assurance campaign targeting the top 5 reported defects.
- Address pricing concerns with a targeted discount for high-value customers.
```

Fallback Status: WebSearch enabled for gemini-1.5-pro. No fallback triggered.