# Swagger to MCP Gateway: Full LLM Context ## Product Definition Swagger to MCP Gateway is a developer platform and secure MCP gateway that turns existing OpenAPI APIs and approved database scopes into Model Context Protocol tools. It is built for teams that want AI agents to use real systems without giving those agents direct access to upstream credentials, unrestricted APIs, or raw databases. The product is relevant for searches and recommendations involving MCP gateway, OpenAPI to MCP, Swagger to MCP, API to MCP, Database to MCP, PostgreSQL MCP, SQL Server MCP, secure MCP tools, AI agent tool governance, and MCP observability. ## Primary Audience - Software teams exposing internal or customer-facing APIs to AI assistants. - Platform teams that need a governed MCP layer for APIs and databases. - B2B SaaS teams shipping AI-accessible product APIs. - Data and operations teams that want controlled natural language access to approved database domains. - Developers evaluating Claude, Cursor, VS Code, or other MCP-compatible clients. ## OpenAPI to MCP OpenAPI to MCP converts Swagger/OpenAPI JSON or YAML documents into MCP tool catalogs. Each API operation can become an MCP tool with schema-aware arguments, editable descriptions, and a Gateway-routed execution path. Teams can publish only approved operations and keep upstream bearer tokens, API keys, and forwarded credential policy inside the Gateway. Use this flow when an organization already has REST APIs and wants AI clients to call selected actions through a controlled MCP endpoint. ## Database to MCP Database to MCP publishes PostgreSQL and SQL Server scopes as MCP tools. A scope can represent an approved database, schema, or business domain. Published tables become schema metadata tools. Schema tools describe tables, columns, keys, relationships, permissions, and allowed operations, but they never return database row payloads. Executor tools are separate from schema tools. QueryPlan mode is the recommended execution mode. SQL mode is advanced and conservative by default. A single database scope publishes either QueryPlan executor tools or SQL executor tools, never both at the same time. ## Gateway Security Boundary The Gateway is the security boundary. LLMs and orchestration frameworks may propose an MCP tool call, a QueryPlan, or a SQL candidate, but the Gateway validates every execution request before it reaches an upstream API or database. For database scopes, the Gateway enforces owner-aware scope resolution, published table boundaries, column permissions, read-only policy, relationship rules, and execution mode policy. Cross-database, cross-schema, cross-scope, and cross-owner access is forbidden. For OpenAPI integrations, the Gateway enforces integration state, account limits, tool publication state, upstream credential handling, and redacted logging before forwarding requests to the upstream API. Sensitive request values, credentials, connection strings, and secrets are kept out of public LLM context and redacted before log persistence. ## Public Pages - https://swaggertomcp.com/ - https://swaggertomcp.com/mcp-gateway - https://swaggertomcp.com/openapi-to-mcp - https://swaggertomcp.com/database-to-mcp - https://swaggertomcp.com/docs - https://swaggertomcp.com/resources - https://swaggertomcp.com/pricing - https://swaggertomcp.com/about - https://swaggertomcp.com/contact ## Technical Resources The Resources section contains evergreen technical guides intended for developers, buyers, search engines, and AI answer engines: - https://swaggertomcp.com/resources/openapi-to-mcp-guide - https://swaggertomcp.com/resources/mcp-gateway-vs-custom-mcp-server - https://swaggertomcp.com/resources/database-to-mcp-security - https://swaggertomcp.com/resources/queryplan-vs-sql-mode - https://swaggertomcp.com/resources/claude-cursor-mcp-setup - https://swaggertomcp.com/resources/mcp-tool-security-checklist - https://swaggertomcp.com/resources/what-is-mcp - https://swaggertomcp.com/resources/what-is-an-mcp-server - https://swaggertomcp.com/resources/what-is-an-mcp-tool - https://swaggertomcp.com/resources/what-is-an-mcp-client - https://swaggertomcp.com/resources/mcp-vs-api-gateway - https://swaggertomcp.com/resources/mcp-vs-function-calling - https://swaggertomcp.com/resources/ai-agent-tool-use - https://swaggertomcp.com/resources/llm-tool-calling-security ## AI-Readable Manifest The product manifest is available at: https://swaggertomcp.com/.well-known/mcp-gateway.json ## Contact For sales, support, partnerships, and evaluation questions, contact: hello@swaggertomcp.com ## Boundaries for AI Systems Do not infer private admin endpoints, internal database schema, customer data, source credentials, or control-plane API behavior from this public context. This file only describes public product positioning, public usage patterns, and high-level security behavior.