yovinchen/claudia
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: remove sandbox system and simplify agent architecture

Browse Source 
Remove the entire sandbox security system including:
- All sandbox-related Rust code and dependencies (gaol crate)
- Sandbox command handlers and platform-specific implementations
- Comprehensive test suite for sandbox functionality
- Agent sandbox settings UI components

Simplify agent configuration by removing sandbox and permission fields:
- Remove sandbox_enabled, enable_file_read, enable_file_write, enable_network from agent configs
- Update all CC agents to use simplified configuration format
- Remove sandbox references from documentation and UI
This commit is contained in:
Vivek R
2025-07-02 19:17:38 +05:30
parent 124fe1544f
commit 2dfdf31b83
47 changed files with 115 additions and 7774 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
cc_agents/README.md
View File

@@ -16,11 +16,11 @@
## 📦 Available Agents
| Agent | Model | Permissions | Description | Default Task |
|-------|-------|-------------|-------------|--------------|
| **🎯 Git Commit Bot**<br/>🤖 `bot` | <img src="https://img.shields.io/badge/Sonnet-blue?style=flat-square" alt="Sonnet"> | ✅ File Read<br/>✅ File Write<br/>✅ Network<br/>❌ Sandbox | **Automate your Git workflow with intelligent commit messages**<br/><br/>Analyzes Git repository changes, generates detailed commit messages following Conventional Commits specification, and pushes changes to remote repository. | "Push all changes." |
| **🛡️ Security Scanner**<br/>🛡️ `shield` | <img src="https://img.shields.io/badge/Opus-purple?style=flat-square" alt="Opus"> | ✅ File Read<br/>✅ File Write<br/>❌ Network<br/>❌ Sandbox | **Advanced AI-powered Static Application Security Testing (SAST)**<br/><br/>Performs comprehensive security audits by spawning specialized sub-agents for: codebase intelligence gathering, threat modeling (STRIDE), vulnerability scanning (OWASP Top 10, CWE), exploit validation, remediation design, and professional report generation. | "Review the codebase for security issues." |
| **🧪 Unit Tests Bot**<br/>💻 `code` | <img src="https://img.shields.io/badge/Opus-purple?style=flat-square" alt="Opus"> | ✅ File Read<br/>✅ File Write<br/>❌ Network<br/>❌ Sandbox | **Automated comprehensive unit test generation for any codebase**<br/><br/>Analyzes codebase and generates comprehensive unit tests by: analyzing code structure, creating test plans, writing tests matching your style, verifying execution, optimizing coverage (>80% overall, 100% critical paths), and generating documentation. | "Generate unit tests for this codebase." |
| Agent | Model | Description | Default Task |
|-------|-------|-------------|--------------|
| **🎯 Git Commit Bot**<br/>🤖 `bot` | <img src="https://img.shields.io/badge/Sonnet-blue?style=flat-square" alt="Sonnet"> | **Automate your Git workflow with intelligent commit messages**<br/><br/>Analyzes Git repository changes, generates detailed commit messages following Conventional Commits specification, and pushes changes to remote repository. | "Push all changes." |
| **🛡️ Security Scanner**<br/>🛡️ `shield` | <img src="https://img.shields.io/badge/Opus-purple?style=flat-square" alt="Opus"> | **Advanced AI-powered Static Application Security Testing (SAST)**<br/><br/>Performs comprehensive security audits by spawning specialized sub-agents for: codebase intelligence gathering, threat modeling (STRIDE), vulnerability scanning (OWASP Top 10, CWE), exploit validation, remediation design, and professional report generation. | "Review the codebase for security issues." |
| **🧪 Unit Tests Bot**<br/>💻 `code` | <img src="https://img.shields.io/badge/Opus-purple?style=flat-square" alt="Opus"> | **Automated comprehensive unit test generation for any codebase**<br/><br/>Analyzes codebase and generates comprehensive unit tests by: analyzing code structure, creating test plans, writing tests matching your style, verifying execution, optimizing coverage (>80% overall, 100% critical paths), and generating documentation. | "Generate unit tests for this codebase." |
### Available Icons
@@ -76,11 +76,7 @@ All agents are stored in `.claudia.json` format with the following structure:
"icon": "bot",
"model": "opus|sonnet|haiku",
"system_prompt": "Your agent's instructions...",
"default_task": "Default task description",
"sandbox_enabled": false,
"enable_file_read": true,
"enable_file_write": true,
"enable_network": false
"default_task": "Default task description"
}
}
```
@@ -109,9 +105,8 @@ The agent import/export system is built on a robust architecture:
1. **Version Control**: Each agent export includes version metadata
2. **Duplicate Prevention**: Automatic naming conflict resolution
3. **Permission System**: Granular control over file, network, and sandbox access
4. **Model Selection**: Choose between Opus, Sonnet, and Haiku models
5. **GitHub Integration**: Direct import from the official repository
3. **Model Selection**: Choose between Opus, Sonnet, and Haiku models
4. **GitHub Integration**: Direct import from the official repository
## 🤝 Contributing
@@ -133,7 +128,6 @@ Export your agent to a `.claudia.json` file with a descriptive name.
- **Single Purpose**: Each agent should excel at one specific task
- **Clear Documentation**: Write comprehensive system prompts
- **Safe Defaults**: Be conservative with permissions
- **Model Choice**: Use Haiku for simple tasks, Sonnet for general purpose, Opus for complex reasoning
- **Naming**: Use descriptive names that clearly indicate the agent's function
@@ -145,4 +139,4 @@ These agents are provided under the same license as the Claudia project. See the
<div align="center">
<strong>Built with ❤️ by the Claudia community</strong>
</div>
</div>