agent-toolbox
🥈SilverAgent-toolbox provides commands, rules, and setups for Claude Code, enabling local development and interoperability with Gemini. It benefits operations teams by streamlining agentic AI workflows and connecting to Xcode for development tasks.
2310Updated 3mo ago
Intermediate30min to implementautomation
Saves ~75 min per use
Quick InstallView Source
git clone https://github.com/czottmann/agent-toolbox.gitWorks with:
Claude
Overview
About This Skill
Agent-toolbox is a collection of configuration files, custom slash commands, and wrapper scripts designed to enhance Claude Code for local development workflows, particularly Xcode-based projects. It includes the claude-wrapper to automatically inject project file lists, reducing token usage and speeding up interactions, plus gemini-wrapper and opencode-wrapper that inherit Claude Code's context for cross-tool consistency. The toolbox provides pre-built rules and MCP server configurations that can be imported into CLAUDE.md files, enabling developers to standardize agentic AI workflows across multiple tools. Setup involves symlinking scripts and commands, then importing rules into project or global configuration files.
How to Use
1. **Install agent-toolbox**: Run `pip install agent-toolbox` or clone from [GitHub repo]. Ensure Python 3.10+ and Xcode (for Swift projects) are installed. 2. **Initialize your project**: Use the prompt template above with your specific project details. The tool will scaffold the project with pre-configured templates for your target language. 3. **Implement features**: Replace [SPECIFIC_FEATURE] with your actual requirements. The tool provides component templates for common agent patterns (routers, API clients, error handlers). 4. **Test locally**: Use `agent-toolbox test` with different modes (unit, integration, end-to-end). The tool runs both language-specific tests and agent-specific validation. 5. **Deploy**: Run the deployment command with your target platform (gemini, claude, or local). The tool handles packaging, dependency resolution, and environment-specific configurations. Check the log file for any warnings or errors.
Use Cases
Accelerate Xcode development with Claude Code by auto-injecting project file lists
Share Claude Code context with Google Gemini for consistent multi-tool workflows
Reduce token usage in build processes with the xcodebuild-wrapper
Set up reusable rules and custom commands for standardized Claude Code interactions
Tags
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/czottmann/agent-toolboxRequirements
- Claude Code or compatible AI agent
- Works with: Claude
Quick Start Guide
1
Install the Skill
Copy the install command above and run it in your terminal.
2
Open Your AI Agent
Launch Claude Code, Cursor, or your preferred AI coding agent.
3
Try It Out
Use the prompt template or examples below to test the skill.
4
Customize
Adapt the skill to your specific use case and workflow.
Usage Examples
Prompt Template
Use the agent-toolbox to [TASK] for [PROJECT_NAME] in [LANGUAGE]. Follow these steps: 1) Initialize the project with `agent-toolbox init --language [LANGUAGE] --project [PROJECT_NAME]`. 2) Implement [SPECIFIC_FEATURE] using the provided templates. 3) Test locally with `agent-toolbox test --mode [MODE]` and ensure compatibility with [TARGET_PLATFORM]. 4) Deploy the agent using `agent-toolbox deploy --target [TARGET_PLATFORM] --env [ENVIRONMENT]`. Document any errors in [LOG_FILE_PATH].
Example Output
### Project Initialization
```bash
agent-toolbox init --language swift --project CustomerSupportAgent
```
Output:
```
Project 'CustomerSupportAgent' initialized in /Users/dev/projects/CustomerSupportAgent
Swift templates loaded: Agent.swift, APIClient.swift, ErrorHandler.swift
Xcode project generated: CustomerSupportAgent.xcodeproj
```
### Feature Implementation
Added a new `TicketRouter` class to handle incoming support tickets:
```swift
class TicketRouter: AgentToolboxComponent {
func route(ticket: Ticket) -> RouteResult {
switch ticket.priority {
case .high: return .immediateEscalation
case .medium: return .assignToTeam(.backend)
case .low: return .scheduleFollowup(days: 3)
}
}
}
```
### Local Testing
```bash
agent-toolbox test --mode integration
```
Output:
```
Running integration tests...
✅ TicketRouter tests passed (5/5)
✅ APIClient tests passed (8/8)
⚠️ Warning: Mock data mismatch in 1 test case
```
### Deployment Preparation
```bash
agent-toolbox deploy --target gemini --env staging
```
Output:
```
Preparing deployment package...
✅ Swift package validated
✅ Dependencies resolved
⚠️ Xcode version 15.2 required (current: 15.1)
📦 Package size: 4.2MB
Deploying to staging environment...
✅ Deployment successful (ID: deploy_20240515_1422)
```Apply to these tools
Browse all tools
Gemini
Google's multimodal AI model and assistant
Claude
AI assistant built for thoughtful, nuanced conversation
Microsoft Teams
Get more done every day with Microsoft Teams – powered by AI
GPT for work
Automate your spreadsheet tasks with AI power
Respell
Agentic AI Workflow platform
Notion
Connected workspace for docs, wikis, and projects
Compatible MCP servers
Browse all MCP servers
Find the right skills for your stack
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan