"OpenAI Gym" matching MCP servers:

Provides tools to manage OpenAI API keys and spending through the OpenAI API. Requires an OpenAI admin API key for secure access to account management features.

Drop-in stdio interceptor that gates MCP tools/call requests through a YAML policy (block / require_approval / audit / budget cap) before they reach the real server. The same policy file is reused by @jamjet/claude-code-hook and @jamjet/openai-guardrail, so one rule set covers Claude Desktop, Cursor, OpenAI Agents, and custom clients.

Secure credential management for AI agents — encrypted storage, auto-refresh, and rate limiting via a single MCP call. Supports OpenAI, Anthropic, Stripe, GitHub, Brave Search, Kalshi, Coinbase, plus generic API key and OAuth2 for any service.

Universal bi-directional rules and prompt transpiler between Cursor (.cursorrules / .mdc), Claude Code (CLAUDE.md), and OpenAI with 98.5% certified AST parity and in-RAM secret redaction.

Any MCP server. Anywhere. An OAuth-protected MCP gateway that can also run the servers, plus your Grok subscription as an OpenAI-compatible API.

Exposes the OpenAI Admin API as MCP tools, enabling management of organization members, projects, service accounts, API keys, rate limits, usage, costs, and audit logs through natural language.

Open-source guardrails between AI agents and FHIR clinical data — PHI redaction, immutable audit, step-up auth, tenant isolation. MCP server + OpenAI/Gemini adapters. A healthclaw.io project.

Enables working with private files on a local machine from ChatGPT through OpenAI Secure MCP Tunnel without opening inbound ports or exposing the local server to the Internet, with tools for listing, reading, creating, modifying, deleting, backing up, and restoring authorized workspace files under strict permissions.

It enables OpenAI models to create and join secure deliberation rooms, post messages, verify Ed25519 W3C DIDs, reach 67% Byzantine-fault-tolerant consensus, manage Atlas 2.0 goal DAGs with rollback, and escalate high-risk actions through fail-closed human approval gates.

Enables read-only, bounded querying of DMARC aggregate report data ingested from an explicitly configured Zoho mailbox folder into SQLite, exposing summary counts, aligned DKIM/SPF failure reporting, per-report details with provenance, coverage gaps and collector health over OpenAI Secure MCP Tunnel. Responses are size-capped and windows are limited to prevent runaway queries, while raw attachments and message bodies are never retained or exposed.

Credential resilience for AI agents, providing AES-256 encrypted storage, automatic health checks, and failover to backup credentials for providers like OpenAI, Anthropic, GitHub, and Stripe.

A security-hardened MCP gateway that enables AI agents to call LLM APIs (Gemini, OpenAI, Claude, etc.) using ephemeral proxy tokens, eliminating exposure of real API keys.

Exposes an OpenAI- and Anthropic-compatible HTTP, SSE, and stdio gateway that wraps multiple subscription CLIs, adding prompt-injection defense, PII redaction, cost-aware routing, and reasoning-trace capture for MCP-compatible clients like Claude Desktop and Cursor.

OpenAI-compatible proxy with PII scrubbing, audit logging, and model routing. Drop-in replacement — set OPENAI_BASE_URL to point to this server. Supports GPT-4o, Claude, Gemini, and Llama via single endpoint.

Turns any OpenAPI 3.0/3.1 spec into validated MCP tools, letting hosts call allowed API operations locally with strict schemas, secure credential handling, and read-only-by-default filtering.

Intent-based security governance for OpenAI Codex and ChatGPT. Register intent plans before tool calls, apply natural-language policy rules, and get per-request audit logs across all your AI agents.

Enables a remote MCP client to access narrowly scoped local filesystem resources and process capabilities through a loopback-only bridge. Supports guarded file reads, searches, and mutations, authenticated transport, and private reachability via OpenAI Secure MCP Tunnel.

Enables AI assistants to perform comprehensive Windows system health diagnostics, security checks, developer workflows, and maintenance tasks via 85 tools with actionable recommendations, over MCP and OpenAI-compatible HTTP.

Open-source AI context firewall for governed model, retrieval, memory, and tool access. It enables OpenAI SDK applications to adopt governed access via a single base_url change and includes MCP support.

Stop runaway agents. One gateway in front of all your agent traffic — MCP tools and LLM models — one budget, one kill switch. Sits between AI clients (Cursor, Claude Desktop, Windsurf) and remote MCP servers / OpenAI-compatible model APIs; hard token budgets; a single static Go binary.

Exposes selected operations of an existing REST/OpenAPI app as MCP tools with generated schemas, permission annotations, and a risk class, requiring explicit opt-in before anything is listed and a second flag for destructive calls. Enforces security-first guardrails such as environment-injected auth, a single confined base URL, timeouts, response caps, rate limiting, and a local test console for trying tools.

Enables interaction with IBM Guardium Cryptographic Manager (GCM) for cryptographic asset management, including authentication, asset inventory queries, policy violation tracking, and ticket management.

A Docker-based relay service that simplifies authentication and provides audit logging for AI agents accessing IBM Guardium Cryptography Manager's MCP server.

An account-scoped MCP connection hub for Claude, OpenAI and Grok, enabling private MCP endpoints, OAuth sign-in, Solana wallet binding, and content publishing via MCP tools.

Implements a secure message communication protocol for handling exchanges between the banking chatbot and Azure OpenAI, providing message queuing, reliability, and detailed logging.

Enables agents to securely use credentials for GitHub, Cloudflare, OpenAI, Stripe, and xAI without ever reading the secret values, including credential health, rotation, and audit features.

Privacy-preserving AI gateway. Sanitises PII before prompts reach Anthropic / OpenAI / your LLM, then emits a signed cryptographic certificate per call (Ed25519 + RFC 3161 + Sigstore Rekor). EU GDPR + AI Act ready. Free tier 500/mo with BYOK.

Let Claude Code, Claude Desktop, OpenAI Codex, Kimi and any MCP-capable AI agent create, launch, check and drive your Nullprint browser profiles. Install with one command: claude mcp add nullprint -- uvx nullprint-mcp

Policy-based governance for AI agent tool calls. YAML policies, approval gates, risk assessment, and audit logging across LangChain, OpenAI, Anthropic, and MCP.

Exposes a single chat tool that lets any MCP client send text prompts together with images, PDFs, or Office attachments to a self-hosted Gemini web endpoint and get OpenAI-compatible responses back. An optional companion checker keeps the underlying Google session cookies fresh automatically so requests keep working without manual re-authentication.