gemini-flow

Overview

Gemini-Flow was a comprehensive AI orchestration platform that transformed how organizations deploy and manage AI systems. It featured 66 specialized AI agents working through A2A and MCP protocols, complete Google AI services integration, and Byzantine fault tolerance for enterprise reliability. Though archived since January 2026, it demonstrated high-performance benchmarks with 396,610 SQLite ops/sec and <75ms routing latency, showcasing sophisticated agent coordination and multi-modal AI workflows.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• veo3_create

  Generate videos using Google's Veo3 AI video generation service

  Try:Create a 60-second product demonstration video with 4K qualityTry:Generate a corporate training video on workplace safety procedures
• imagen4_create_batch

  Generate high-fidelity images in batches using Google's Imagen4 service

  Try:Create 4 professional headshots for LinkedIn profilesTry:Generate batch of marketing banners for social media campaign
• lyria_compose

  Compose background music using Google's Lyria AI music generation

  Try:Create a 60-second corporate upbeat background trackTry:Compose professional energetic music for a product video
• chirp_synthesize

  Generate voiceovers using Google's Chirp text-to-speech service

  Try:Create a professional female voiceover for a product introductionTry:Generate a male voice narration for an educational video
• jules_remote_create

  Execute development tasks remotely using Jules VM with quantum optimization

  Try:Implement OAuth 2.0 authentication with quantum optimizationTry:Create a REST API endpoint for user authentication
• jules_local_execute

  Execute development tasks locally using agent swarm with quantum optimization

  Try:Refactor a monolithic application to microservices architectureTry:Optimize database queries for better performance
• hive_mind_spawn

  Deploy coordinated agent teams for enterprise solutions

  Try:Spawn agent team for enterprise digital transformation projectTry:Create a swarm of specialized agents for code review and optimization
• auth_setup

  Configure authentication with Google AI services

  Try:Setup authentication with Google AI services using service accountTry:Configure API credentials for Google Cloud Project
• agents_spawn

  Spawn coordinated agent teams for intelligent coordination

  Try:Spawn 20 specialized agents for intelligent coordinationTry:Create a swarm of agents for data processing and analysis
• monitor

  Monitor A2A coordination and system performance in real-time

  Try:Show me A2A coordination metrics and performance benchmarksTry:Display agent swarm status and task completion rates
• co_scientist_research

  Conduct research using Google's Co-Scientist AI research assistant

  Try:Research latest developments in quantum computingTry:Analyze market trends for renewable energy sector
• mariner_automate

  Automate data extraction tasks using Google's Project Mariner

  Try:Extract data from financial reports and create summariesTry:Automate invoice processing and information extraction

Note: Inferred tools from CLI commands and API examples in the README. Some signatures were not explicitly provided.

Comparable tools

Installation

npm install -g @clduab11/gemini-flow
gemini-flow init --protocols a2a,mcp --topology hierarchical

gemini extensions install https://github.com/clduab11/gemini-flow
gemini extensions enable gemini-flow

FAQ

Why was the project archived?

The author archived the project on January 29, 2026, stating that generative AI is moving too fast and that some of Google's latest releases (Antigravity) discontinued the need for this type of project.

Can I still use Gemini-Flow?

The project is in read-only/archived state, so you can explore the code and documentation, but it's no longer actively developed or maintained.

Compare gemini-flow with