lobehub/lobehub
LobeHub is a platform that enables multi-agent collaboration and human-agent co-evolution by treating AI agents as units of work, providing tools to build, manage, and collaborate with AI agent teams seamlessly.
Awesome MCP › Agent Orchestration Platforms
eschnou/OpenGPA
OpenGPA is a self-hosted, open-source general-purpose agent written in Java on top of Spring AI. Rather than being a chat front end over a model, it implements an agentic loop: the model acts as the brain of an agent that plans across multiple steps, reasons verbally about what to do next using the ReAct approach, picks an action from a catalogue, executes it and observes the result before continuing. Each step is emitted as a structured object holding the reasoning text, the chosen action with its arguments and a flag for whether it is final, and the bundled user interface surfaces that internal reasoning so an operator can see why the agent did what it did. The action catalogue is extensible in code, so an integrator can add actions that call internal services, APIs or a workflow engine; the shipped set includes web search and web browsing through Playwright, artifact upload and download, retrieval-augmented search over an imported document store, and Model Context Protocol support that lets the agent reach existing enterprise APIs exposed as MCP servers. Because it builds on Spring AI it works with the major model providers, including Anthropic and OpenAI as well as locally run LLama or Mistral models, which supports the project's offline-first stance that data need not leave the operator's own server. Deployment is by Docker Compose: a quickstart file brings up PostgreSQL with the pgvector extension, the backend and the frontend together, and building from source needs Java 21 and Maven. The user interface lives in a separate opengpa-frontend repository, and a REST API allows programmatic use. Prompt and interaction logging can be switched on for debugging. The roadmap lists persistent memory, code generation and execution, scheduled jobs, triggers, task persistence, token-cost auditing and observability. MIT licensed.
https://github.com/eschnou/OpenGPA
Also in Agent Orchestration Platforms
tinyhumansai/openhuman
OpenHuman is a local-first personal AI assistant with persistent Markdown memory, multi-agent workflow orchestration, and research tools, connecting to 100+ OAuth integrations and thousands of MCP servers.
assafelovic/gpt-researcher
GPT Researcher is an autonomous AI agent that conducts deep local and web research on any topic, generating detailed, unbiased reports with citations and supporting integration with the Model Context Protocol (MCP) for enhanced data sourcing.
deepset-ai/haystack
Haystack is an open-source AI orchestration framework for building production-ready LLM applications in Python, enabling modular pipelines and agent workflows with explicit control over retrieval, ...
activepieces/activepieces
Activepieces is an open-source AI automation platform providing an extensible MCP toolkit and over 280 MCP servers for AI agents, designed for both technical and non-technical users.
openai/swarm
Swarm is an educational framework by OpenAI for lightweight, scalable multi-agent orchestration using agents and handoffs, designed to explore and teach multi-agent coordination techniques.
superset-sh/superset
Superset is a desktop application designed to orchestrate and manage multiple CLI-based AI coding agents in parallel, using isolated Git worktrees for efficient, context-separated development.
VoltAgent/voltagent
VoltAgent is an open-source TypeScript framework for building and orchestrating autonomous AI agents with support for the Model Context Protocol (MCP) to enable standardized tool interactions and scalable multi-agent systems.