AGIFlow
Agiflow provides a shared project board for people and external AI assistants. It keeps projects, tasks, comments, files, decisions, and progress in one durable record so ChatGPT, Claude, Cursor, Codex, and other MCP-compatible clients can work from the same scoped project state.
At a Glance
- Solo builders
- Small software and product teams
- Freelancers and client-service teams
- Marketing teams
- +3 more
AI Tools by AGIFlow
(1)Agiflow
Project Board for AI Assistants
Discussions
No discussions yet
Be the first to start a discussion about AGIFlow
Latest News
OpenAI DevDay 2026: Choose What Your Team Delegates to ChatGPT
Freelance client file handover: check who controls each file
Typed Decisions for AI Agent Workflows: Record Before Routing
AgiFlow aicode-toolkit release activity reached @agiflowai/style-system 0.2.1 alongside multi-client plugin updates
Products & Services
A web-based project board where humans and approved external AI assistants manage projects, work units, tasks, statuses, comments, artifacts, decisions, and workflow locks through scoped MCP access.
An iPhone project-planning and work-tracking app for reviewing project state, tasks, comments, artifacts, workflow history, and decisions; it also provides on-device Agent chat.
An open-source toolkit for AI-assisted development, including project and feature scaffolding, file-level design guidance, rule-based review, design-system discovery, and multi-client agent plugins. Licensed AGPL-3.0.
An MCP server that generates projects and features from reusable templates and approved project patterns.
Market Position
Agiflow positions itself as a focused project-board and shared-state layer for external AI assistants rather than as an AI model, an autonomous agent host, or a traditional diagram workflow engine. Its differentiation is durable, scoped, human-reviewable project state across multiple assistants; its public comparisons identify Trello, Asana, and Jira as conventional alternatives, while its content discusses agent-oriented alternatives such as Vibe Kanban and MCP project-management tools.
Leadership
Founders
Vuong Ngo
Full-stack developer and the public builder/author behind Agiflow.io; he maintains the AgiFlow/aicode-toolkit GitHub repository and publishes technical writing on AI-assisted development. Public profiles also describe him as building Agiflow.io and associate him with Eased UP.
Executive Team
Vuong Ngo
Founder and builder
Public Agiflow author, contact, GitHub maintainer, and full-stack developer focused on AI-assisted development and MCP tooling.
Founding Story
Agiflow was created around the problem that AI-assisted work becomes fragmented across ChatGPT, Claude, Cursor, Codex, terminal sessions, and team chats. Its initial vision is to make a project board the durable authority layer: assistants do the work through approved MCP tools, while the board retains scope, tasks, decisions, artifacts, locks, ownership, and review state.
Business Model
Revenue Model
SaaS subscriptions for the Agiflow project-board platform, priced by seat, with included remote-execution allowances and prepaid credits for additional remote execution. The open-source aicode-toolkit is distributed under AGPL-3.0.
Pricing Tiers
Up to 2 seats, 3 projects, 2 API keys, 3 AI-assistant connections, and 30 minutes of remote execution per month.
Unlimited seats, projects, API keys, and assistant connections, plus 300 remote-execution minutes per seat per month.
Charged from prepaid credit balances after included minutes; credit packs are $10, $25, $50, or $100, and usage stops at the stated balance floor rather than creating surprise over-billing.
Target Markets
- Solo builders
- Small software and product teams
- Freelancers and client-service teams
- Marketing teams
- Sales teams
- Business owners and team leads
- AI-assisted software development and coding-agent handoffs
- Marketing campaign planning and execution
- Freelance client projects and deliverable tracking
- Sales follow-up pipelines
- Small-team project management
- Reviewing and approving agent-produced work