Pentest MCP: Professional Penetration Testing Toolkit

@DMontgomery40

リポジトリ

基本情報

カテゴリ

開発者ツール

ライセンス

MIT

トランスポート

stdio

作者

DMontgomery40

Pentest MCP: Professional Penetration Testing Toolkit について

NOT for educational purposes: An MCP server for professional penetration testers including STDIO/HTTP/SSE support, nmap, go/dirbuster, nikto, JtR, hashcat, wordlist building, and more.

#mcp #cybersecurity #nmap #pentesting#jtr#john-the-ripper#nikto#dirbuster#redteam#pentesting-tools#gobuster#model-context-protocol #mcp-server

Pentest MCP: Professional Penetration Testing Toolkit の設定

標準の設定はありません

このサーバーの README には解析可能な MCP 設定ブロックが含まれていません。インストール手順はリポジトリをご確認ください。

リポジトリ

Pentest MCP: Professional Penetration Testing Toolkit のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Pentest MCP: Professional Penetration Testing Toolkit?

A professional penetration-testing MCP server that integrates with common security tools (nmap, John the Ripper, hashcat, gobuster, nikto, subfinder, httpx, ffuf, nuclei, hydra, sqlmap, tcpdump) and provides an agent-friendly interface for reconnaissance, exploitation, reporting, and engagement management. It targets penetration testers and red teams who want to automate tool orchestration via MCP.

How to use Pentest MCP: Professional Penetration Testing Toolkit?

Install globally with npm install -g pentest-mcp. Run locally over stdio with pentest-mcp, or over the network with MCP_TRANSPORT=http MCP_SERVER_HOST=0.0.0.0 MCP_SERVER_PORT=8000 pentest-mcp. Use pentest-mcp inspector to launch the bundled MCP Inspector. Tools are invoked via MCP tool calls (e.g., subfinderEnum, nucleiScan). Auth can be enabled with bearer tokens and OIDC when using HTTP transport.

Key features of Pentest MCP: Professional Penetration Testing Toolkit

20+ penetration testing tools including nmap, John the Ripper, hashcat, gobuster, nikto, subfinder, httpx, ffuf, nuclei, hydra, sqlmap, tcpdump
Modern transport support: stdio (default), Streamable HTTP (recommended), and deprecated SSE
Bearer-token authentication with optional OIDC JWKS and introspection support
Engagement record tracking and automated report generation with scope-of-work (SoW) handling
Built-in MCP Inspector launcher – no separate install required
Docker deployment with all tools pre-installed
Supports interactive and template-based SoW capture for client reports

Use cases of Pentest MCP: Professional Penetration Testing Toolkit

Automate subdomain enumeration and live host probing for external network assessments
Fuzz web content paths, run template-based vulnerability scanning, and test for SQL injection
Perform brute-force checks on services (SSH, etc.) and capture network traffic for analysis
Generate post-engagement reports with structured engagement records and scope-of-work notes

よくある質問

What transport modes are supported?

stdio is the default for local MCP clients; Streamable HTTP (set `MCP_TRANSPORT=http`) is the recommended network transport. SSE is available only as a deprecated compatibility mode and will be removed in a future major release.

How do I enable authentication?

Set `MCP_AUTH_ENABLED=true` and `MCP_AUTH_MODE=bearer` when using HTTP (or deprecated SSE) transport. Configure OIDC with `MCP_OIDC_ISSUER`, `MCP_OIDC_JWKS_URL`, and optionally `MCP_OIDC_INTROSPECTION_URL`. Legacy aliases (`MCP_OAUTH_ENABLED`, etc.) are temporarily accepted.

What tools must be installed locally for non-Docker runs?

Ensure these binaries are in `PATH`: `nmap`, `john`, `hashcat`, `gobuster`, `nikto`, `subfinder`, `httpx-toolkit` (or validated ProjectDiscovery `httpx` as fallback), `ffuf`, `nuclei`, `hydra`, `sqlmap`, `tcpdump`.

How do I provide a scope of work for a report?

Use `createClientReport` with `scopeMode=ask` (elicits from user interactively), `scopeMode=provided` (paste SoW text directly into `scopeOfWork`), or `scopeMode=template` (uses a built-in generic authorized-testing template). Elicitation is recommended; if unavailable, the template is used as fallback.

Is this server safe for unpermissioned use?

No. The README states "Authorized use only. Run against systems/networks where you have explicit written permission." Prompt injection risks are documented, and mitigations like least privilege, allowlists, and human confirmation for destructive actions are recommended.

「開発者ツール」の他のコンテンツ

Moxie Docs

Jackalope-Dev

Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.

FARPY

mangomunchr

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP

DrUrls

DrUrls audits websites for SEO, and its MCP server lets Claude, ChatGPT, Claude Code, Cursor or VS Code work with those audits directly: list your sites with their latest score and grade, read each issue across SEO, perf

UXMachine

UXMachine measures a web page in a real browser and returns verifiable observations, each with its evidence and its limits. It opens the address on desktop (1280×800) and mobile (390×844) and follows the site's own links

Mcp-Warden

ermoluk

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.