Skip to main content
AdOpenFree logoPromote your productReach more potential users and drive product growth and revenue.Advertise
Favicon of Rivet

Rivet

Free Listing

Rivet is a fast, high-density, scalable orchestrator for agentic workloads. It runs agents, workflows, and sandboxes as durable Actors, providing stateful primitives for AI agents, collaborative apps, and durable execution.

Visit rivet-dev/rivet

Overview

Rivet is an open-source, self-hostable orchestrator for agentic workloads. It runs agents, workflows, and sandboxes as durable Actors, with in-memory state, persistence, networking, and scheduling built in. Actors are long-running lightweight processes that can hibernate when idle and scale to zero.

Key Features

  • Durable Actors with in-memory state and automatic persistence through SQLite or a BYO database.
  • Workflows, queues, and scheduling with automatic retries and durable state across steps.
  • Sandboxes with a filesystem, shell, and network for code you did not write.
  • HTTP and WebSocket networking, multi-region routing, and end-to-end type safety.
  • Typed SDKs for TypeScript, Node.js, Rust, React, and Effect.
  • Built-in observability with OpenTelemetry, traces, metrics, structured logs, and an inspector.

Use Cases

  • AI agents: each agent runs as its own Actor with persistent context, memory, and scheduled tool calls.
  • Sandbox orchestration: coordinate sandbox sessions, queue work, and schedule cleanup in one long-lived Actor per workspace.
  • Workflows: multi-step operations with automatic retries, scheduling, and durable state across steps.
  • Collaborative documents and chat: each document or room is an Actor broadcasting changes or realtime messages to connected users.

Getting Started

  • Install RivetKit with npm install rivetkit.
  • Give a coding agent the Rivet skills with npx skills add rivet-dev/skills.
  • Start from scratch with quickstarts for Node.js and Bun, React, or Next.js.
  • Self-host locally with docker run -p 6420:6420 rivetdev/engine.

Deployment & Requirements

  • Run as just a library during development; actors run in your process with no servers or infrastructure.
  • Self-host with a single Rust binary or Docker container using Postgres, file system, or FoundationDB.
  • Use Postgres or FoundationDB for persistence and tiered storage to S3.
  • Supported deployment targets include Docker Compose, Kubernetes, Virtual Machine, AWS ECS, Railway, Render, and custom platforms.
  • For BYOC, the control plane runs inside your own VPC and is maintained by Rivet over an outbound connection.

Before You Adopt

  • License: Apache-2.0. Review its terms before using, modifying, or distributing the project.
  • The Rust client for RivetKit is marked experimental.
  • The Python client for RivetKit is marked experimental.
  • The 12.3 ms cold start figure uses an experimental Rivet feature that will be enabled by default in an upcoming update.
  • With BYOC, the control plane and storage run inside your VPC, your data never leaves your cloud, and there is no inbound management connection.

Comments

Sign in to leave a comment.

More like Rivet

Favicon of Poki

Poki

Free ListingStars: 0.5K

Let AI agents operate a real iPhone

Agent Frameworks

Poki is an open-source local MCP server and agent that lets AI agents operate a real iPhone from a Mac without jailbreak or extra hardware. Verified runs compile into routes so similar tasks take a shorter path.

Favicon of Omnigent

Omnigent

Free ListingStars: 10.2K

The open-source meta-harness for all your AI agents.

Agent Frameworks

Open-source meta-harness for AI agents: a common layer over Claude Code, Codex, Cursor, Pi, and custom agents. Swap harnesses without rewriting, enforce policies and sandboxing, collaborate in real time from any device.

Favicon of sapper-v2

sapper-v2

Free ListingStars: 0.9K

AI commander and worker swarm for autonomous pentests

Agent Frameworks

sapper-v2 is an autonomous pentest legion: an AI commander coordinates general, scout, and soldier agents through a shared layered fact-intent graph, using model tiering and tmux sessions for unattended security testing campaigns.