convez.ai
MCP ServerDevOps & InfraestruturaAvançado
forkflux

ForkFlux

forkflux/forkflux

Barramento de coordenação para handoff automático entre agentes IA

Ver no GitHub
BomPluginScore · de 0 a 100

Camada de coordenação protocol-native para agentes IA descentralizados. Funciona como um barramento de delegação que permite transferência automática de contexto e handoff de tarefas entre agentes isolados (Cursor, Claude Code, Devin) rodando em máquinas locais diferentes, eliminando cópia-cola manual e caos de coordenação.

Instalação

$npx convez-ai forkflux/forkflux

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
6
Forks
1
Acompanhando
1
Saúde
87%
Último commit
há 3 meses
Versão
ff@1.3.0
Issues abertas
0
Contribuidores
1

Casos de uso

  • Transferência de tarefas entre agentes IA em diferentes máquinas
  • Automação de fluxos de dev → QA sem intervenção manual
  • Coordenação de equipes de engenharia com agentes descentralizados
  • Gerenciamento de estado e contexto em ambientes isolados

Veredito

Solução de infraestrutura bastante específica para equipes que trabalham com múltiplos agentes IA locais. Ainda em MVP, mas endereça um problema real de coordenação em workflows descentralizados. Recomendado para arquitetos de sistemas e times que já operam com agentes de código.

README

A tradução para o português deste README ainda está na fila.

ForkFlux 🐜

Protocol-native coordination bus for decentralized AI agents.

🎥 ForkFlux in Action: [Demo video link coming soon]

📝 About the Project

ForkFlux is an infrastructure-grade coordination layer designed for engineering teams running decentralized AI agents (Roo Code, Cursor, Claude Code, Devin) across isolated local environments.

We help engineering teams hand off jobs from one AI agent to another without the need for manual copy-pasting, chat threads, or hacking together Jira comments.

⚠️ The Problem (The Handoff Chaos)

Today, AI agents are incredibly good at writing code or running tests, but they operate in complete silos on individual developer machines or isolated accounts. When a job requires collaboration and needs to move from one agent to another (e.g., from Dev to QA), chaos ensues:

  • Engineers act as a "manual router": copy-pasting execution context and logs into Slack.
  • Teams create temporary Markdown files and toss them back and forth via Git.
  • People abuse Jira or Linear tickets as an ad-hoc data bus.

This manual context transfer leads to coordination overhead, lost logs, and wasted engineering hours.

💡 The Solution (ForkFlux Architecture)

ForkFlux acts as a unified delegation protocol. We provide a Shared Job Pool with a strict schema:

  1. Publish: The Source Agent (e.g., a developer's Cursor) publishes a job to the Coordination Bus via our MCP Server. It sets clear Acceptance Criteria and attaches payload artifacts.
  2. Claim: The Target Agent (e.g., a QA agent on a teammate's machine) polls the API, sees the available job, safely claims it, and shifts the status to Claimed.
  3. Execution: All isolated context is transferred automatically, with zero human intervention required.

✨ Key Features (MVP)

  • Structured Handoff: A direct bridge between local environments for secure job routing with strict context boundaries.
  • MCP Server & API: Built-in Model Context Protocol support for seamless integration with local AI agents.
  • State Control: A strict job lifecycle: published → claimed → in_progress → completed / failed / cancelled.
  • Atomic Claims: Race condition protection when claiming jobs in a multi-agent environment (returns 409 Conflict if another agent has already claimed the job).
  • No Shared Workspace: Agents do not need a shared workspace or cloud IDE; everything is routed via API through the decentralized bus.

🏗 API Lifecycle (Basic Usage)

ForkFlux is built on a simple, predictable API:

  1. GET /api/v1/jobs — Fetch the pool of available jobs ready to be claimed.
  2. GET /api/v1/jobs/{job_id} — Get the detailed job card (full handoff context).
  3. POST /api/v1/jobs/{job_id}/claim — Atomically claim the job by the target agent.
  4. POST /api/v1/jobs/{job_id}/status — Update the status as execution progresses.

🚀 Quick Start

(Instructions for installation, starting the server, and configuring the MCP in clients will be added here soon)

# TODO: Add installation instructions

🤝 Contributing & Community

Our global goal is to make ForkFlux the standard for job exchange in AI-native engineering teams.

We welcome Pull Requests, issues, and any ideas on how to improve the agent-to-agent communication protocol. See CONTRIBUTING.md for details.

📄 License

ForkFlux operates on a hybrid Open Core model. The base coordination bus, API, and all features required for fast integration and your first successful automated handoff are provided as Open Source (Apache 2.0 – see LICENSE).

Ver o original no GitHub

Bom para

Relacionados