rivetkit
0RivetKit provides persistent, in-memory actor processes for AI agents and multiplayer applications. Suitable for developers needing durable execution and real-time communication capabilities without the overhead of traditional databases.
1700Updated 2mo ago
AI Agent Skill
Quick InstallView Source
$ npx skills add https://github.com/rivet-dev/skills --skill rivetkitOverview
About This Skill
RivetKit is a backend framework and Rivet Actor runtime for building high-performance, long-lived in-memory processes. It's designed for AI agents, real-time collaboration, workflow orchestration, and geo-distributed data scenarios where HTTP endpoints, databases, or message queues fall short. Developers use RivetKit to build, debug, and deploy actor-based systems with durable execution and real-time communication capabilities. The framework includes built-in debugging tools via inspector HTTP APIs, support for serverless and runner modes, and deployment guidance for production environments. It's suited for teams building stateful AI agents and multiplayer applications requiring low-latency, persistent state management.
How to Use
Install with the command: `$ npx skills add https://github.com/rivet-dev/skills --skill rivetkit`
Use Cases
Build AI agents with persistent state management
Create multiplayer applications without relying on traditional databases
Automate complex workflows with durable execution patterns
Tags
Setup & Installation
Quick Install
Terminal
$ npx skills add https://github.com/rivet-dev/skills --skill rivetkitAlternative Install (Git Clone)
git clone https://github.com/rivet-dev/skillsRequirements
- Claude Code or compatible AI agent
Quick Start Guide
1
Install the Skill
Copy the install command above and run it in your terminal.
2
Open Your AI Agent
Launch Claude Code, Cursor, or your preferred AI coding agent.
3
Try It Out
Use the prompt template or examples below to test the skill.
4
Customize
Adapt the skill to your specific use case and workflow.
Usage Examples
Prompt Template
Design a multi-agent system for [COMPANY] in the [INDUSTRY] sector using RivetKit. The system should include [NUMBER] persistent actors with these roles: [ROLE_1], [ROLE_2], and [ROLE_3]. Each actor should maintain state for [TIMEFRAME] and communicate via [COMMUNICATION_METHOD]. Include error handling for [POTENTIAL_FAILURE_SCENARIO].
Example Output
### Multi-Agent System Design for GreenTech Solutions
**System Overview:**
A persistent actor system for real-time energy grid optimization with 3 actors:
- **GridMonitor**: Tracks power consumption and forecasts demand spikes
- **ResourceAllocator**: Dynamically adjusts renewable energy distribution
- **AlertHandler**: Triggers emergency protocols during outages
**Actor Specifications:**
**1. GridMonitor (Actor ID: gm-001)**
- Persistence: 7-day rolling state cache
- State: {current_load: 1245.8 MW, forecast: [1320, 1410, 1530], last_updated: "2024-05-15T14:30:00Z"}
- Communication: WebSocket push to ResourceAllocator every 30s
**2. ResourceAllocator (Actor ID: ra-002)**
- Persistence: 24-hour state retention
- State: {solar_active: 68%, wind_active: 42%, battery_level: 74%}
- Communication: REST API calls to GridMonitor and AlertHandler
**3. AlertHandler (Actor ID: ah-003)**
- Persistence: Event-driven state updates
- State: {active_alerts: [], resolved_alerts: ["outage-2024-05-15-001"]}
- Communication: SMS/Email notifications for critical events
**Error Handling:**
- Network failures: 3-retry exponential backoff
- State corruption: Automatic snapshot recovery from last 30 minutes
- Actor crashes: Supervisor restarts within 5s with previous stateFind the right skills for your stack
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan