ruflo
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive m
What is ruflo?
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
Overview of ruflo
What it is
Key features
Use cases
Getting started
License and pricing
Ruflo is an agent meta-harness for Claude Code and Codex. It provides an execution layer that adds tools, memory, loops, sandboxes, and controls to AI agents. One command (npx ruflo init) gives Claude Code a system where agents self-organize into swarms, learn from tasks, remember across sessions, and communicate securely across machines without leaking data. The architecture follows a conceptual flow: User → Ruflo (CLI/MCP) → Router → Swarm → Agents → Memory → LLM providers.
Key features
- 100+ specialized agents, coordinated swarms, self-learning memory, federated comms across machines, and enterprise security guardrails.
- Adaptive memory, self-learning intelligence, federation, vector RAG integration, and native integration with Claude Code, Codex, Hermes, and others.
- Visual console (agent cockpit) that runs inside Claude Code or standalone; displays Overview, Cost, Swarm, Events, Missions, and an optional AI page.
- Plugin marketplace with installable mods, skills, and agent definitions; supports /plugin commands to add, install, enable, disable, and update.
- MCP server for connecting Ruflo tools to any MCP-enabled client.
- Hooks system that automatically routes tasks, learns from successful patterns, and coordinates agents in the background.
Use cases
- Deploying intelligent multi-player swarms for coordinated autonomous workflows.
- Building conversational AI systems with persistent memory across sessions.
- Inspecting agent logs, triaging failures, replaying and comparing runs.
- Using Claude Code with enhanced routing, policy enforcement, and cost tracking.
Getting started
Path A — Claude Code Plugins (lite, slash commands only):text/plugin marketplace add ruvnet/ruflo/plugin install ruflo-core@ruflo/plugin install ruflo-swarm@ruflo/plugin install ruflo-rag-memory@ruflo/plugin install ruflo-neural-trader@rufloPath B — CLI install (full Ruflo loop):bashnpx ruflo@latest init wizardFor MCP clients, configure command npx and arguments `[
Based on the project's public README and repository metadata.