"Local RAG system for providing documentation to a large language model (LLM)" matching MCP servers:

MCP server toolkit with payload library, WAF bypass generator, rate limiter, plugin system, and inter-MCP bridge for security testing.

An MCP server that runs curated adversarial prompts against local Ollama models to test guardrails, scoring responses with heuristic verdicts for human review.

MCP server that guides LLM security reviews step-by-step, tracking progress and ensuring all assigned source code is read before completing each plan step.

Lets any MCP client, such as Claude, browse attack suites and probes, run a full assurance assessment with all reports written out, send custom prompts or multi-turn histories and get leak, PII and refusal checks back for adaptive attacks, and read findings from an earlier assessment. This enables agent-driven red-teaming of a chatbot, RAG app or AI agent without leaving the chat client.

Enables AI agents to perform offline cybersecurity research and penetration testing by querying a local knowledge base of curated security data, with tools for searching, answering, and status checking.

Two-way Burp Suite MCP bridge enabling AI agents to capture traffic, analyze endpoints, queue scans, and send findings back to Burp.

Enables AI coding agents to audit websites for PHP runtime vulnerabilities, compile a live Zero-PHP Next.js App Router replacement preview from a WordPress/PHP site, and retrieve remediation guidance.

Enables Claude Code to leverage OpenAI models for expert code review, deep architecture analysis, and security audits, with automatic markdown report generation.

Wraps Kali Linux penetration testing tools for authorized security assessments, driven by a local LLM via LM Studio.

GDB Enhanced Features MCP server for remote debugging and CTF exploitation with 41 specialized tools including ROP search, format-string detection, and memory patching.

Enables MCP clients to run hardened, read-only radare2-based static binary triage tools, including disassembly, cross-reference following, string decoding, function renaming, and evidence-checked report generation without executing the sample.

Connects AI assistants to a Kali Linux terminal for executing security tools, managing interactive shell sessions, and performing penetration testing, CTF challenges, and security research tasks.

Provides 180+ tools for full system control including terminal, Git, Docker, Kubernetes, VMs, BIOS, kernel, cybersecurity, AI, and cloud, designed to work with MCP clients like Cursor and Claude Desktop.

Enables LLMs to perform automated penetration testing and Active Directory reconnaissance through Mythic, with tools for executing PowerShell, AD recon, domain user enumeration, and Kerberoasting.

Exposes a hardened Docker container with Kali Linux security tools (nmap, sqlmap, dig, whois, etc.) as MCP tools, enabling network reconnaissance, web analysis, and vulnerability scanning through natural language commands.

An MCP server that scans local codebases for quantum-vulnerable cryptography (secp256k1, Ed25519, RSA, etc.) and CI signing commands, classifying each finding as quantum-broken, post-quantum, or neither. It runs entirely locally with no network calls, providing a deterministic inventory for AI agents.

Enables LLM agents to browse and triage vulnerability observations, manage products, branches and rules, import scan reports and SBOMs, run scans and background jobs, and generate VEX documents via SecObserve's REST API.

SkillTotal is a local stdio MCP server that lets an agent check a component before installing it. It statically analyzes MCP servers, agent skills, npm/PyPI packages and repos for supply-chain risks, dangerous capabilities (shell, filesystem, network, credentials), prompt-injection surfaces like tool poisoning, and exfiltration paths. Each finding cites file, line and snippet.

Enables AI agents to access VirusTotal intelligence through MCP, supporting file, URL, domain, and IP lookups, plus analysis, via remote HTTP or local stdio.

MCP security firewall (stdio): vet advertised tool definitions through static-scan → threat-feed → origin → pinning before they reach the model. Zero npm runtime deps. No secrets.