AI clients are best for planning and coordinating work through conversation. Coding agents are best for implementing scoped tasks. Both use the same Agiflow board, so decisions, task status, and delivery context stay connected.

Integration	Type	Best for
ChatGPT	AI client	Plan and update project work from ChatGPT
Claude	AI client	Refine plans and tasks from Claude
Cursor	Coding agent	Keep coding work aligned with the project board
Codex	Coding agent	Turn scoped board tasks into delivered code
Claude Code	Coding agent	Run terminal-based coding work from shared tasks
Antigravity	Coding agent	Plan and implement work with Google Antigravity
Any MCP-compatible client	AI client	Connect a custom MCP-compatible workflow

Learn how shared AI project context works

These practical guides explain how MCP connections, shared project boards, and coding-agent context fit together.

Use one project board across ChatGPT, Claude, Cursor, and Codex Compare MCP project management tools Stop AI coding agents from losing project context Connect Claude Code to a shared project board

Agiflow integration questions

Which AI tools integrate with Agiflow?

Agiflow has dedicated integrations for ChatGPT and Claude, plus support for any client that can connect through the Model Context Protocol.

Which coding agents can use an Agiflow project board?

Cursor, Codex, Claude Code, and Antigravity can use an Agiflow board as shared project context for planning and delivery.

What does an Agiflow integration let an AI tool do?

A connected AI tool can work from the tasks and project context you authorize, then read or update that shared board instead of relying on pasted summaries.

Can I connect a custom AI client or agent?

Yes. Agiflow supports the Model Context Protocol, so you can connect a custom MCP-compatible client by following the connection guide.

Build your own integration

Agiflow speaks the Model Context Protocol, so any MCP-compatible client can read and update your board.

Read the docs