Cybersec Toolkit MCP server
{
"mcpServers": {
"26zl-cybersec-toolkit": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/26zl/cybersec-toolkit:1.1.1"
],
"env": {
"SUDO_USER": "<sudo_user>",
"GITHUB_TOKEN": "ghp_..."
}
}
}
}{
"mcpServers": {
"26zl-cybersec-toolkit": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/26zl/cybersec-toolkit:1.1.1"
],
"env": {
"SUDO_USER": "<sudo_user>",
"GITHUB_TOKEN": "ghp_..."
}
}
}
}{
"mcpServers": {
"26zl-cybersec-toolkit": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/26zl/cybersec-toolkit:1.1.1"
],
"env": {
"SUDO_USER": "<sudo_user>",
"GITHUB_TOKEN": "ghp_..."
}
}
}
}{
"servers": {
"26zl-cybersec-toolkit": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/26zl/cybersec-toolkit:1.1.1"
],
"env": {
"SUDO_USER": "${input:sudo_user}",
"GITHUB_TOKEN": "${input:github_token}"
}
}
},
"inputs": [
{
"id": "sudo_user",
"type": "promptString",
"description": "Sudo user",
"password": true
},
{
"id": "github_token",
"type": "promptString",
"description": "Github token",
"password": true
}
]
}{
"mcpServers": {
"26zl-cybersec-toolkit": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/26zl/cybersec-toolkit:1.1.1"
],
"env": {
"SUDO_USER": "<sudo_user>",
"GITHUB_TOKEN": "ghp_..."
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 1.2.1
What it does
An MCP server that exposes a registry of 670+ security tools for CTF, penetration testing, and digital forensics. It allows AI agents to discover tools, check installation status, recommend tools for specific security tasks, and execute them through a governed execution path. Designed for security professionals and CTF competitors who want AI-assisted tool discovery and execution.
- Query the tool registry to find applicable security tools for a specific CTF category or vulnerability type.
- Run installed security tools through the MCP server with argument validation and execution governance, without invoking a shell directly.
- Get tool recommendations from an AI agent based on the target type or security problem at hand.
Tools this server exposes
| Tool | Does |
|---|---|
| query_tool_registry | Search and list available security tools in the registry by name, category, or profile. |
| check_install_status | Verify whether specific security tools are installed and ready to use. |
| recommend_tools | Get AI-suggested tools for a given CTF category, vulnerability type, or security task. |
| run_tool | Execute an installed security tool with argument validation and network policy enforcement. |
| run_pipeline | Execute a sequence of security tools in order, with output passed between steps. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.