AgentGPT and agentok

These two tools are competitors, as both aim to provide a browser-based, visual interface for assembling and deploying autonomous AI agents, with AgentGPT being a more established and widely adopted solution.

AgentGPT
51
Established
agentok
48
Emerging
Maintenance 0/25
Adoption 10/25
Maturity 16/25
Community 25/25
Maintenance 2/25
Adoption 10/25
Maturity 16/25
Community 20/25
Stars: 35,812
Forks: 9,436
Downloads:
Commits (30d): 0
Language: TypeScript
License: GPL-3.0
Stars: 399
Forks: 63
Downloads:
Commits (30d): 0
Language: TypeScript
License:
Archived Stale 6m No Package No Dependents
Stale 6m No Package No Dependents

About AgentGPT

reworkd/AgentGPT

🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.

Implements goal-directed agent loops using LangChain to recursively decompose tasks, execute them via external APIs (OpenAI, Serper, Replicate), and refine based on results. Built on Next.js 13 + FastAPI with Prisma/SQLModel ORM, MySQL backend, and Next-Auth authentication—supporting both cloud and self-hosted deployments via Docker.

About agentok

dustland/agentok

AG2 Visualized - Build Multi-Agentic Apps with Drag-and-Drop.

Agentok Studio helps you design and manage complex AI-driven workflows by letting you visually arrange different AI agents and define how they interact. You drag and drop components to create a blueprint, and the tool generates the underlying Python code. This is for AI solution architects, developers, or researchers who need to build advanced multi-agent applications without deep coding for every step.

AI-workflow-design multi-agent-systems AI-application-development prompt-engineering AI-orchestration

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