arcwall-security MCP Server

The io.github.rom-baro/arcwall-security MCP server provides security scanning for AI coding tools. It detects secrets, produces threat models, and runs pre-commit checks. It is described as applicable to Claude Code, Cursor, Windsurf, and “any MCP-compatible AI coding tool.” Tooling includes specific scan capabilities.

🛠️ Key Features

• Detects hardcoded secrets
• Generates threat models
• Runs pre-commit checks
• Includes a tool named arcwall_scan_secrets (hardcoded credentials)

🚀 Use Cases

• Scanning AI coding tool workflows (Claude Code, Cursor, Windsurf, other MCP-compatible tools)
• Pre-commit security checks for AI-assisted development
• Secret detection for repositories using MCP integrations

⚡ Developer Benefits

• Centralizes security scanning across MCP-compatible AI coding tools
• Provides a dedicated secrets scanning tool: arcwall_scan_secrets

⚠️ Limitations

• Setup requires an API key from https://arcwall.io and configuration in the MCP settings for the target AI tool