Unla and mcp-gateway

These are competitors offering different architectural approaches to MCP gateway functionality: Unla emphasizes zero-code transformation of existing servers with lightweight Docker deployment, while lasso-security/mcp-gateway prioritizes plugin-based orchestration and enterprise agent development patterns.

Unla
54
Established
mcp-gateway
43
Emerging
Maintenance 16/25
Adoption 10/25
Maturity 9/25
Community 19/25
Maintenance 10/25
Adoption 10/25
Maturity 9/25
Community 14/25
Stars: 2,051
Forks: 169
Downloads:
Commits (30d): 2
Language: TypeScript
License: MIT
Stars: 352
Forks: 28
Downloads:
Commits (30d): 0
Language: Python
License: MIT
No Package No Dependents
No Package No Dependents

About Unla

AmoyLab/Unla

🧩 MCP Gateway - A lightweight gateway service that instantly transforms existing MCP Servers and APIs into MCP servers with zero code changes. Features Docker deployment and management UI, requiring no infrastructure modifications.

Supports dynamic configuration hot-reloading via OS signals, HTTP, or Redis PubSub, enabling runtime updates without service restarts. Built in Go for minimal resource overhead, it bridges RESTful APIs and existing MCP servers to MCP-compliant endpoints using YAML-driven configuration, with multi-tenant session management and OAuth pre-authentication support. Includes persistent storage options (SQLite, PostgreSQL, MySQL) and native Kubernetes/Helm deployment support alongside Docker.

About mcp-gateway

lasso-security/mcp-gateway

A plugin-based gateway that orchestrates other MCPs and allows developers to build upon it enterprise-grade agents.

Supports pluggable security guardrails (basic token masking, PII detection via Presidio, and advanced policy enforcement via Lasso) that intercept and sanitize MCP requests/responses before reaching LLMs. Exposes `get_metadata` and `run_tool` endpoints to provide unified discovery and execution across proxied MCPs, while optionally scanning server reputation via Lasso Security's API. Integrates with Cursor and Claude Desktop through standard `mcp.json` configuration, with Docker support for containerized deployment.

Scores updated daily from GitHub, PyPI, and npm data. How scores work