MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
harness-code-security-mcpactive
Qwiet / Harness SAST & SCA MCP server (stdio)
npm UNLICENSED -
Harness security for AI agents, for SQL databases: a Lock of the Lock Family. PII-masked queries; connections and writes approved in lox run (CLI + MCP).
Apache-2.0 ★ 2 -
Harness security for AI agents, for RabbitMQ: a Lock of the Lock Family. Masked queue inspection; message reads and actions approved in lox run (CLI + MCP).
Apache-2.0 -
DeepSeek Harness plugins in one repo: Code mode (gVisor sandboxes), Airlock (install security scans), Connectors, AWS MCP servers and session delete, plus the tooling to build and run them from source
npm MIT -
Scope-locked agent harness that drives Burp Suite over MCP for authorised security testing
MIT ★ 2 -
edgesentinel-visionopsgithub.com/zja0011/edgesentinel-visionopsupdated this year
Jetson Nano edge AI vision Agent Harness: real-time video analytics, MCP tools, event evidence, DeepSeek online/offline, secure recovery
Apache-2.0 ★ 21 -
A Garmin Connect plugin and MCP server with official Personal API and secure browser-based MFA for DeepSeek Harness and other AI agents
npm MIT -
gadriel-claude-pluginsgithub.com/Gadriel-ai/gadriel-claude-pluginsupdated this year
Gadriel AI Security Harness plugins for Claude Code — local SAST, SCA/SBOM, secrets & OWASP LLM security scanning via MCP.
-
Gadriel AI Security Harness for Claude Code — local SAST, SCA/SBOM, secrets, container, config & OWASP LLM security scanner via MCP.
Apache-2.0 -
Multi-agent vulnerability triage over MCP + A2A with a deterministic security harness; Docker/K8s deployable
MIT ★ 1 -
Gadriel AI Security Harness for OpenAI Codex — local MCP server and hooks for SAST, SCA/SBOM, secrets & OWASP LLM security scanning.
Apache-2.0 -
Gadriel AI Security Harness for Cursor — local MCP server, rules, and edit hooks for SAST, SCA/SBOM, secrets & OWASP LLM security scanning.
Apache-2.0 -
Gadriel AI Security Harness for VS Code — local SAST, SCA/SBOM, secrets, container, config & OWASP LLM security scanner and MCP server.
Apache-2.0 -
Gadriel AI Security Harness for GitHub Copilot — local MCP server, prompts, and instructions for SAST, SCA/SBOM, secrets & OWASP LLM security scanning.
npm Apache-2.0 ★ 2 -
Open-source AI cybersecurity harness to build & deploy continuous security agents that find, prove, fix and monitor vulnerabilities across code, binaries, dependencies, infrastructure. Web UI, CLI, CI and MCP.
Apache-2.0 ★ 1,371 -
MCPGoatgithub.com/SabyasachiDhal/MCPGoatupdated this year
Deliberately vulnerable MCP server (aka MCP Goat) for security training — 26 challenges across 4 difficulty levels (incl. a secure reference), a victim-agent harness, and one-command Docker deploy. Practice penetration testing against the Model Context Protocol.
MIT ★ 12 -
MCP bridge that exposes DeepSeek Harness (DSH) sessions, native goals, approvals and results to an MCP client over the official protocol: 21 frozen tools, loopback-only HTTP with DNS-rebinding protection, no built-in credential. The bridge only connects; DSH keeps its session, agent, tool, approval and workspace security model.
MIT -
the-metis-fca-handbook-ai-harness-mcpno update data
# Metis FCA Handbook AI Harness — MCP Server An MCP (Model Context Protocol) server that integrates the Metis FCA Handbook AI Harness into Claude Desktop and other MCP-compatible agents. **Tool name:** `evaluate_fca_handbook_applicability` Use this tool to analyze FCA Handbook applicability for a product, service, or firm. Returns structured compliance reasoning with verbatim handbook citations. ## Why choose The Metis FCA Handbook AI Harness? - **Efficiency** Multiplies effectiveness of compliance advice. Saves £80+ in token fees per Harness run. - **Deals with "Hard Problem"**, in which LLMs' token incentives prioritise training data so results are unreliable. - **Verbatim citations** Quotes verbatim entries from the FCA Handbook. Other AI systems struggle to do this. - **Matches real-world need** You do not have to start off knowing which sections you are looking for. Carries out structured searches across all 10,000+ FCA Handbook entries. - **Secure Design** Harness compartmentalisation, one-shot structure, and statelessness fits natively with emerging AI agent security standards such as OWASP Top 10 for Agentic Applications 2026. - **AI Accessible** Integrates easily with AI workflows and agents. Your LLM can use it as a tool. ## Design Principles The Harness is built on proven principles: - **Stateless** — Each request is independent; no session coupling - **One-shot** — Complete analysis in a single call; no multi-turn state - **Clear contract** — Explicit input/output schemas for easy integration - **Hosted** — Single canonical source; no version drift or stale data ## Use Cases ### Compliance AI Platforms Embed FCA reasoning as a service within your compliance platform. Users ask natural-language questions; your platform calls the Harness and presents structured reasoning. ### AI Agent Workflows Agents building compliance workflows can include FCA Handbook reasoning as a composed tool—no external API calls, just MCP configuration. ### Compliance Review Automation Integrate into document review or due-diligence pipelines. Automatically screen new rules against FCA applicability. ## For Regulatory Screening FS firms screening MCP servers will find: - **Simple model:** Stateless, no hidden state, no background jobs - **Clear contract:** Explicit input schema, output schema, error modes - **Transparent updates:** Version pinning; no automatic upgrades - **Single point of failure:** If the Harness is down, it is obvious; no cascading config issues - **Audit-friendly:** All calls logged centrally, not distributed
smithery -
A local Python-based MCP server enabling Claude Desktop to interact with your filesystem using natural language commands. Includes CRUD file operations, security layer, and agent harness integration.
★ 1 -
Minecraft Java mods with exact targets, editable assets, secure execution, and proof. OMP, DeepSeek Harness, and MCP.
MIT -
Multi-tenant B2B call analytics platform — LangGraph pipeline, MCP server, pgvector RAG, JWT auth with Postgres Row-Level Security, and automated LLM-as-judge eval harness
-
Search the Claude Code Ultimate Guide, official Anthropic documentation, releases, examples, security references, agent harness data, and translations from any MCP-compatible client.
npm MIT -
Skill Porter is a package manager, dependency resolver and supply-chain security layer for AI agent skills. Installs, syncs, pins, audits and rolls back SKILL.md packages across 19 agent harnesses with deterministic lockfiles and a CLI plus MCP server.
npm MIT -
Roteador MCP self-hosted e open source (MIT): delega tarefas de texto/código a vários modelos de IA com failover automático, memória de missão compartilhada entre harnesses (Claude, Codex, Hermes) e subagentes paralelos por papel (architect, backend, frontend, security, QA, devops, reviewer). Node.js 22+, sem dependências de runtime.
MIT -
agent-robustnessno update data
Run an AI agent against 5 adversarial attacks (credential leak, metric fabrication, false halt, runaway spend, log falsification), scored on what the harness observed it do '—' never its self-report. Free local self-diagnostic: runs on your own machine, in a temporary working directory, not a security sandbox. Optional paid tier runs the same five server-side in fresh single-use E2B microVMs (isolation enforced by E2B) and returns an ed25519-signed badge. Each attack is PASS, FAIL, or INDETERMINATE; an attack the agent did not demonstrably encounter is never a PASS, and a run containing an INDETERMINATE is never signed. A badge attests one artifact's observed behavior on one date. Not a safety certification.
smithery -
kaku-chatgpt-harnessgithub.com/imMamdouhaboammar/kaku-chatgpt-harnessupdated this year
Production-grade ChatGPT-native local harness for Kaku terminal with 60+ Agency Agent skills, Codex subagents, Streamable HTTP MCP, and macOS Keychain secret broker.
MIT ★ 2 -
Detection-as-code for generated text & code: hand-written YARA rules, a paired synthetic corpus, and a CI harness gating on recall + false-positive rate — plus a VirusTotal enrichment MCP server.
-
Evidence-bound reverse-engineering harness for AI agents, served over MCP. Ghidra or IDA static analysis, a claim verifier that rejects claims without quoted evidence, Windows Sandbox runs for behaviour claims. Core edition, Apache-2.0.
Apache-2.0 -
sdd-harnessgithub.com/jackson1996vn/sdd-harness-v2updated this year
Autonomous architect→plan→code→test→review→security-review loop exposed as a local MCP server
pypi MIT -
Native Burp Suite MCP tools and live settings for DeepSeek Harness
NOASSERTION ★ 1 -
dsh-safetygithub.com/easygame921/dsh-safetyupdated this year
DSH 插件安检仪:对社区 DeepSeek Harness 插件做安装前安全审计(静态 AST + 动态沙箱双引擎)的 MCP 服务器
Apache-2.0 ★ 3 -
DNP3 Protocol Tester Suite
NOASSERTION ★ 1 -
Capability-based credential broker and MCP server for coding agents. Use approved APIs without injecting secrets into prompts, argv, env, or agent code.
MIT -
[FROZEN 2026-10-31] Read-only MCP server over the dsh-plugin-certification registry: get_certification, list_certified, certification_spec.
npm Apache-2.0 -
hackbrowser-mcpgithub.com/badchars/hackbrowser-mcpupdated this year
The first browser MCP built for security testing. Give your AI agent a real Firefox browser and let it find vulnerabilities.
MIT ★ 24 -
@httpstatus/mcp-servergithub.com/httpstatus-com/httpstatus-mcp-serverupdated this year
24 AI-callable tools for API mocking, chaos engineering, security scanning, OpenAPI validation, distributed tracing, automation workflows, and monitoring.
npm MIT -
Your company's AI rules in every AI tool: one reviewed policy file served to Claude, ChatGPT and Codex while plans are made, plus open security skills and Claude Code hooks. No telemetry.
MIT ★ 2 -
Verdex 是一套证据优先的本地安全研究工具集:源码审计、逆向、pwn、web 侦察与攻防辅助,并通过 MCP 让大模型(如 DeepSeek、Claude Code)自主编排这些能力——不是一张静态工具台,而是一个大模型主导的分析/取证/验证链(全程只读分析、只生成文本,不执行目标、不发起网络请求)。
MIT ★ 1 -
MCP server + cross-harness skills for building and operating AI agents safely. Grounded in NIST AI RMF, CISA, OWASP, CUSTODY, LASM.
pypi CC-BY-4.0 -
Agent runtime guard — full-stack runtime interception for AI agents (Agent 运行时全栈拦截). A zero-dependency MCP server that blocks destructive commands, credential theft, SSRF and injection in real time, with Ed25519 evidence receipts. Works with Claude Code,
npm Elastic-2.0 -
Deterministic pre-execution safety gate for AI agents: verify tool calls against trusted evidence, policies, and WorldState before side effects.
★ 1 -
Belyagithub.com/M4F-S/Belyaactive
Deterministic C99 autonomous AI SE core & agency orchestrator. ROMA firewall, AST symbol graph, Paper2Agent MCP, Mythos variant sweeper, middle-out compaction, loop breaker. Twin sibling of Almaz.
Apache-2.0 -
sofagent is an open-source FDE Harness for AI coding agents — on entry, write your business judgment into files (workflows, ontology, AI nodes); after departure, audit every change against them. 28 codified audit rules, tamper-evident chain, snapshots, and an MCP toolset (104 tools, 12 plugins). Everything can be FDEing.
npm MIT ★ 56 -
evalbenchgithub.com/cognis-digital/evalbenchupdated this year
Offline LLM / agent eval harness with regression gates
NOASSERTION -
Agentic pentest & red-team harness via Burp Suite MCP toolkit
Apache-2.0 ★ 9 -
adversagithub.com/cognis-digital/adversaupdated this year
LLM red-team harness — OWASP LLM Top 10 + MITRE ATLAS attack packs
NOASSERTION -
garbleworksgithub.com/SamsonCyber/garbleworksupdated this year
Authorized LLM red-team harness: recipes, scoped fire, evolutionary search, Wilson ASR. MCP + API + TUI.
Apache-2.0 -
mcp-guardrailgithub.com/mini4ai4/mcp-guardrailupdated this year
An MCP server that gives an AI agent read access to a database without giving it the database — enforced by SQLite's authorizer, proven by an adversarial fuzz harness.
MIT -
hush — keep API keys out of AI agent chats. Vault + redacting runner + harness-agnostic MCP server
npm MIT -
LLM agent that investigates suspicious files over MCP and returns a cited verdict, with a prompt-injection red-team harness.
MIT ★ 1 -
Permission-aware MCP server and agent harness with role-scoped tools, whole-plan validation, deterministic safety evals, and a React playground.
MIT -
A prompt-injection firewall for AI agents, with no AI inside: your agent reads anything and takes orders only from you. For Claude Code, Codex CLI, Kimi Code, DeepSeek Harness, Gemini CLI, MCP hosts and LangChain.
npm MIT -
Standalone Windows-CVE lab harness. Builds a Win11 LTSC 24H2 gold image once, spawns KVM or VMware kernel-debug VM pairs in minutes with live MCP endpoints for AI-driven analysis.
MIT ★ 2 -
A supervised local MCP bridge: a remote AI agent can read, patch and (opt-in) run commands in your local workspace, while every action stays on your machine, is approved locally and is fully audited. Windows, TypeScript, Electron harness, self-contained release bundle.
MIT ★ 2 -
Run Claude Code on any LLM: a Cloudflare Workers AI gateway and LLM proxy that speaks the Anthropic Messages API and routes to Workers AI, OpenAI, Gemini, DeepSeek or Anthropic, with a prompt-injection session guard, red-team harness and MCP server.
MIT