ИИ-агенты
mcp-gateway-registry
An enterprise MCP gateway: governed, auditable tool access for agents.
★ 964 звездыApache-2.0
Смотреть руководство →open-connector — проект с открытым исходным кодом из категории «ИИ-агенты» от oomol-lab (oomol-lab/open-connector на GitHub). У него 5 968 звёзд на GitHub и 545 форков, он распространяется по лицензии Apache-2.0 и последний раз обновлялся на GitHub 8 октября 2026.
An open-source auth gateway connecting 1,500+ SaaS providers to AI agents via SDK, CLI, MCP and HTTP.
List the providers (GitHub, Gmail, Notion, BigQuery and more) and the actions your agent must perform on each.
Start Open Connector with Docker Compose, which pulls the published GHCR image.
Add a provider connection — start with a simple personal-access-token provider like GitHub, or configure OAuth2 — in the web console.
Run a no-auth Action to verify the runtime, then call a credentialed Action from your agent via MCP, SDK or HTTP.
A working connector catalog across products such as GitHub, Gmail, Notion, BigQuery and Google services, with consistent provider and Action IDs.
Agents receive metadata, safe account labels and execution results — provider secrets, scopes and tokens stay inside the inspectable runtime.
The same Actions are exposed through an SDK, a CLI, MCP, plain HTTP and OpenAPI, so any agent stack can use them.
Supports API keys, OAuth2, custom credentials and no-auth providers, with token refresh and named connections.
Connection identity, scope controls, runtime tokens, action allow/block policies and run logs are all managed in the runtime and web console.
Deploy locally with Docker/Node (SQLite or PostgreSQL) or use OOMOL's hosted runtime, keeping the same contracts.
Clone Open Connector and enter the project directory.
git clone https://github.com/oomol-lab/open-connector.git
cd open-connectorBring the runtime up with Docker Compose (this pulls ghcr.io/oomol-lab/open-connector:latest).
docker compose upOpen the local console, connect a provider, and run a no-auth Action to verify the runtime.
# Open http://localhost:3000 — see docs/quickstart.md for the first connection walkthroughOpen Connector is maintained by oomol-lab (OOMOL).
The gateway is Apache-2.0 licensed and free to self-host; OOMOL also offers a hosted runtime for teams that prefer not to run it.
The gateway itself needs no key; each SaaS provider you connect requires that provider's own credentials (API key or OAuth app), which the gateway stores for you.
All three connect agents to SaaS tools; Open Connector's emphasis is a self-hostable gateway where credentials, scopes, policies and logs stay inside your own runtime boundary.
Clone the repo, run docker compose up, open http://localhost:3000, and connect GitHub with a personal access token.
An enterprise MCP gateway: governed, auditable tool access for agents.
Open-source hub of integrations, SDK and CLI for building and deploying GPT/LLM chatbots and agents.
An MCP server and CLI that lets AI agents build, test and run iOS and macOS projects.
Mobile app automation and verification built for AI coding agents — CLI, MCP server, and Node API
Bridge AI assistants to the Unity Editor via MCP: manage assets, scenes, scripts and tasks.
MCP aggregator, orchestrator, middleware and gateway in one Docker — compose many servers into one endpoint
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если open-connector вам не подходит.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Соседние проекты из категории «ИИ-агенты» в каталоге, ближайшие по числу звёзд к open-connector — продолжайте изучать полку по обе стороны от него.
The benchmark that grades AI on real GitHub issues.
A framework for evaluating and improving AI agents.
A desktop-grade AI agent from NetEase Youdao that does real work — analysis, slides, docs, video and research.
A zero-code platform for auto-generating production-grade AI agents with unified tools, skills and memory.
Open-source AI coworkers, each with a computer of their own.
Python SDK for AI agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks including CrewAI, Agno, OpenAI …
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.