kali-mcp-tool
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| kali_execute_commandB | Executes a shell command directly on the Kali Linux host system. Args: command: The shell command line to execute (e.g. 'nmap -sS 192.168.1.1', 'nikto -h http://example.com') timeout_seconds: Maximum time in seconds before terminating the command (default: 300s) cwd: Working directory path for execution (default: ./kali_workspace) |
| kali_search_toolsA | Searches installed Kali Linux tools by name, keyword, or category. Categories include: 'information-gathering', 'vulnerability-analysis', 'web-application-analysis', 'password-attacks', 'wireless-attacks', 'reverse-engineering', 'exploitation-tools', 'sniffing-spoofing', 'post-exploitation', 'forensics' |
| kali_get_tool_helpA | Retrieves complete manual (manpage) or --help documentation for any Kali Linux tool. Args: tool_name: Name of the binary utility (e.g. 'nmap', 'sqlmap', 'john', 'aircrack-ng') |
| kali_run_nmapB | Runs Nmap network scanner against target hosts with options. Args: target: Target IP address, hostname, or subnet (e.g. '192.168.1.1' or 'scanme.nmap.org') options: Nmap flags (default: '-sV -F' fast service version scan) |
| kali_run_sqlmapA | Runs Sqlmap automatic SQL injection and database takeover tool. Args: target_url: Target URL to test (e.g. 'http://testphp.vulnweb.com/artists.php?artist=1') options: Sqlmap flags (default: '--batch --banner' non-interactive banner scan) |
| kali_run_gobusterB | Runs Gobuster URI/directory brute-forcer. Args: target_url: Target URL (e.g. 'http://192.168.1.100') wordlist: Path to wordlist file (default: '/usr/share/wordlists/dirb/common.txt') options: Additional gobuster dir flags (e.g. '-x php,html,txt') |
| kali_run_searchsploitA | Searches Exploit-DB offline database via Searchsploit. Args: query: Search keywords (e.g. 'Apache 2.4', 'WordPress 5.0', 'Linux Kernel 5.4') |
| kali_manage_workspaceA | Manages files inside the local Kali workspace folder. Args: action: 'list' (list files), 'read' (read file content), 'write' (create/overwrite file), 'delete' (remove file) filename: Relative filename in workspace content: Text content (required when action is 'write') |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
The generic kali_execute_command overlaps with the dedicated run_nmap, run_sqlmap, run_gobuster, and run_searchsploit wrappers, since the same operations can be performed through either path. However, the tool descriptions clarify intended use, and the other tools are functionally distinct.
All tools consistently use the kali_ prefix and follow a clear verb_noun pattern such as execute_command, search_tools, run_nmap, and manage_workspace. This makes the toolset predictable and easy to navigate.
Eight tools is a well-scoped size for a Kali Linux MCP server: generic execution, tool discovery, help retrieval, several common security tool wrappers, and workspace file management cover the main use cases without unnecessary bloat.
The set covers the core Kali workflow of finding tools, reading documentation, running common scanners, searching exploits, and managing workspace files. The generic command executor fills many long-tail gaps, though there is no dedicated handling for results post-processing or less common tool categories.