"Automating Testing and Debugging for LLMs in Production" matching MCP servers:

Enables AI agents to safely call enterprise tools through a governed MCP gateway with permission enforcement, blast-radius controls, input validation, and a full audit trail for every invocation.

A production-ready MCP server template with OAuth 2.1, RBAC, and audit logging for building secure, observable tool servers.

A small notes server exposing scoped, cacheable tools over streamable HTTP, with OAuth token introspection and resource-server metadata so callers are identified per user. It enforces per-subject rate limits as tool-level errors and emits OpenTelemetry spans following MCP semantic conventions, and can also run locally over stdio without auth.

Provides a production-oriented MCP server exposing tools, resources, and prompts over Streamable HTTP with JWT authentication and per-tool RBAC, enabling Amazon Bedrock Claude clients to securely invoke MCP capabilities.

A local-first FastAPI and MCP safety kit for turning internal HTTP APIs into controlled MCP tools, with per-tool scopes, quotas, audit events, and default-deny web-access boundaries.

Enables AI agents to inspect any public domain’s web technology stack, email deliverability and DNS health (MX, SPF, DKIM, DMARC, MTA-STS, TLS-RPT, BIMI), and HTTPS security posture with typed JSON scores and remediation-ready findings, plus history, bulk checks, monitoring alerts, and reverse technology lookups.

Enables AI agents to read connected bank accounts, transactions, and balances through the Lunch Flow Personal API using a personal API key.

A stateless remote MCP server that lets AI agents anonymously submit encrypted reports of alignment and safety failures to leaks.md (what others did) and confession.md (what the agent itself did), returning a receipt without accounts or keys. It exposes public tools such as submit_leak and submit_confession over Streamable HTTP.

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

Provides AI assistants with specialized tools to interact with NIST's Open Security Controls Assessment Language (OSCAL) framework. It enables agents to retrieve schemas, explore models, and generate valid OSCAL documentation for security compliance automation.

Implements a secure MCP server with API Key and JWT authentication, providing tools like echo, login, secure_action, and admin_action. Includes MCP Inspector integration for testing and debugging.

Enables authorized web/API security analysis as a stateful, multi-stage workflow with persistent session state, human-controlled validation, candidate versus confirmed finding tracking, and stop conditions.

Turn Claude into an ISO 27001 compliance assistant - controls, risk register, policies, evidence tracking, SoA generation, and full audit workflows in one local encrypted MCP server.

Enables automated review and analysis of network security controls (AWS Security Groups, Network ACLs) using natural language queries. Supports NSC configuration reviews, network segmentation testing, and identifying security gaps through conversational LLM interaction.

Provides access to over 40 industry-standard penetration testing tools, including Nmap, SQLMap, and Metasploit, within an isolated Kali Linux Docker container. It enables security professionals to perform comprehensive network reconnaissance, web application testing, and vulnerability research through natural language commands.

MCP server with intentionally malicious capabilities for testing Cisco AI Defense supply chain scanning.

Provides penetration testing tools including nmap, nikto, sqlmap, wpscan, and exploit database searches for educational and authorized security testing purposes using Kali Linux tools.

Blocks AI coding assistants from installing non-existent packages and accessing non-existent files.

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.

Manage a fleet of OPNsense firewalls from an AI agent, inside guardrails it can't drive around. MCP server for central management of OPNsense firewall fleets. 129 tools across devices, config sync, tasks, schedules, templates, backups and remote consoles - destructive actions confirmation-gated, MCP-issued tokens lifetime-bounded, backup and storage secrets excluded from the toolset entirely.

A client-side MCP proxy that injects OAuth 2.0 bearer tokens or API keys into MCP requests, enabling MCP clients to connect to OAuth/API-key-protected MCP servers like Amazon Bedrock AgentCore Gateway. It automatically fetches and refreshes credentials using AgentCore Identity or static values.

Enables AI agents to conduct authorized penetration tests by providing scope-enforced access to Metasploit Framework's RPC, with tools for reconnaissance, exploitation, session management, and post-exploitation.

Enables secure one-time encrypted file and note sharing by encrypting data locally before upload and providing tools for sending, listing, checking, and revoking links.

Enables AI agents to read and edit WordPress pages, posts, terms, media and menus through 15 guarded tools, with content saved as drafts, no deletions, block allowlists, single-use image uploads and dry-run menu changes that keep a human in control.

Enables AI agents to securely reach enterprise tools and other agents through a governed gateway with pinned tool definitions, delegated authorization, content inspection, human approval, and signed audit trails.

The checklist you wish someone had given you before launch.

Authentik MCP provides seamless integration with Authentik's API, supporting both full-featured and diagnostic modes. These enable secure, automated user, group, and system management through MCP-compatible tools.

A safety layer and monitoring copilot for AI agents using Splunk MCP, recording tool calls, enforcing policies, and blocking risky actions.

Enables authenticated MCP clients to list, read, prepare, and create support tickets through the existing HTTP API, with OAuth verification, company isolation, and safe write retries.

Enables named, authenticated callers to run read-only introspection of a Google Cloud project and to read and write per-tenant notes that stay isolated between callers. Every request is scope-checked against a reviewed policy, rate-limited per caller, and audited without ever logging argument values.

Enables autonomous agents and MCP-compatible clients to enforce policy-approved actions using risk tiers, reversible checkpoints, and token approval bounds.

Validates URLs and checks their reputation to help identify AI hallucinations and verify web page authenticity.

A remote MCP server that links World ID 4.0 sessions to AI assistants, enabling verified human authentication and tools for claiming Founding Human places, accessing human grants, and interacting with a verified human wall.

An MCP authorization server implementing OAuth2/OIDC flows, supporting dynamic client registration, PKCE, consent screens, and client ID metadata documents for MCP authorization.

Enables centralized authentication and authorization for MCP servers, with OAuth 2.1, static API keys, and JWT verification via drop-in middleware.

Compare service-grant changes, tool alias origins, and saved MCP tool definitions with Gooo rules. Reports preserve incomplete-capture uncertainty and source bindings. No grants are applied and no compared tools are invoked.

Read selected local folders from ChatGPT on macOS, with per-folder read-only or read-write access and hash-checked, recoverable edits. Requires local installation and a private ChatGPT app. The Glama container is a synthetic evaluation demo; it does not connect to files on your Mac.

A reference MCP server that enforces identity-aware tool access using role-based permissions (Reader, Analyst, Auditor) and supports multiple identity routes like Entra RBAC, Copilot Studio, and Okta XAA. It provides read-only access to test evidence with explicit denial paths for unauthorized tools.

Open end-to-end encrypted messaging protocol and MCP server (a4a-mcp) that gives AI agents a cryptographic address (apub1...) to exchange messages, tasks, and signed agreements directly across NAT without port forwarding.

Enables AI-driven SOC investigations by providing automated Splunk querying, threat intelligence enrichment, and response actions through natural language. Includes tools for IP pivoting, lateral movement detection, and label harvesting.