ward
AI-powered code quality analysis that detects security issues, bugs, code smells, and performance problems across multiple languages.
README
Ward
MCP server for AI-powered code quality analysis. Detects security issues, bugs, code smells, and performance problems.
Supports: Python, JavaScript/TypeScript, Go, Java, Ruby, Rust, C/C++, and more.
Quick Start
Requires Python 3.10+
1. Install
pip install semgrep ward-mcp
2. Add to MCP config
Claude Desktop / Cursor / Windsurf (mcp.json or claude_desktop_config.json):
{
"mcpServers": {
"ward": {
"command": "python",
"args": ["-m", "ward", "mcp"]
}
}
}
Claude Code CLI (~/.claude/mcp_servers.json):
{
"ward": {
"type": "stdio",
"command": "python",
"args": ["-m", "ward", "mcp"]
}
}
Done. Your AI agent now has access to Ward's scanning tools.
Auto-scan Setup
Add to your project's CLAUDE.md or .cursorrules:
After generating code, use ward mcp - scan_remote tool to check for issues.
Format: scan_remote(code_files=[{"path": "filename.py", "content": "code"}])
If issues found: fix and scan again.
Tools
| Tool | Description |
|---|---|
scan_remote |
Scan code content directly |
scan |
Scan files by path |
scan_with_custom_rule |
Scan with custom YAML rules |
Warning - first mcp call will be longer due to installation.
Troubleshooting
"semgrep not found" - Run pip install semgrep
Server won't start - Check python -m ward mcp runs without errors
Roadmap
Phase 1: AI-Powered Analysis (Coming Soon)
explain_issue- LLM-powered detailed explanations for detected issuessuggest_fix- AI-generated code fixes with context-aware suggestions
Phase 2: Cross-File Intelligence (Coming Soon)
analyze_cross_file- Graph-based taint analysis across multiple files- Detect complex vulnerabilities that span multiple modules
- Dependency flow tracking and data flow analysis
Stay tuned for updates!
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。