Plugin Overview
TokenRing Coder provides 37+ extensible plugins that add powerful capabilities to your AI-powered development workflow. Each plugin integrates seamlessly with the agent framework to provide specialized functionality.
🎯 Core Plugins
Essential plugins for agent functionality and AI integration.
Plugin | Description |
---|---|
Agent | Agent and AgentTeam management with tools, commands, and hooks |
AI Client | Unified interface for multiple AI providers (OpenAI, Anthropic, Google, etc.) |
CLI | Command-line interface utilities and argument parsing |
Memory | Short-term memory and attention storage for agents |
History | Conversation history management and persistence |
💾 Storage & Database
Plugins for data persistence and database operations.
Plugin | Description |
---|---|
Database | Abstract database interface for multiple providers |
MySQL | MySQL database integration with connection pooling |
SQLite Storage | SQLite-based checkpoint storage for agent state |
Checkpoint | Checkpoint management system for saving agent states |
Vault | Secure encrypted vault for managing secrets and credentials |
📁 Filesystem
File system operations and implementations.
Plugin | Description |
---|---|
Filesystem | Abstract filesystem with operations and ignore filters |
Local Filesystem | Concrete local disk filesystem implementation |
S3 | AWS S3 filesystem and CDN implementations |
File Index | File indexing and search with semantic chunking |
Code Watch | File watching and change detection for live updates |
🔧 Development Tools
Plugins for code development and testing.
Plugin | Description |
---|---|
Git | Git integration with auto-commit and branch management |
Testing | Testing framework with auto-repair hooks |
JavaScript | JavaScript/TypeScript utilities and operations |
Codebase | Codebase analysis and indexing capabilities |
Scripting | Scripting engine integration for custom workflows |
Template | Reusable AI-powered prompt templates for common tasks |
☁️ Cloud & Infrastructure
Cloud service integrations and container orchestration.
Plugin | Description |
---|---|
AWS | AWS service integrations and utilities |
Docker | Docker integration via CLI with sandbox provider |
Kubernetes | Kubernetes orchestration and management |
Sandbox | Abstract sandbox interface for isolated execution |
CDN | Abstract CDN service interface for content delivery |
🌐 Web & Search
Web scraping, search, and browser automation.
Plugin | Description |
---|---|
Web Search | Abstract web search with pluggable providers |
Serper | Serper.dev API integration for Google search |
ScraperAPI | ScraperAPI integration for web scraping and SERP |
Wikipedia | Wikipedia API integration for knowledge retrieval |
Chrome | Chrome browser automation via Puppeteer |
Reddit integration for searching posts and retrieving content | |
Research | Web-enabled AI research using Gemini models |
🎵 Audio & Media
Audio recording, playback, and transcription.
Plugin | Description |
---|---|
Audio | Abstract audio framework for recording and playback |
Linux Audio | Linux-specific audio implementation with naudiodon2 |
🛠️ Utilities
General-purpose helper plugins.
Plugin | Description |
---|---|
Utility | Promise handling, caching, logging, and shell escaping |
Queue | Task queue management for async operations |
Tasks | Task scheduling and execution framework |
Feedback | User feedback collection and management |
Iterables | Iterable utilities and providers |
📝 Content & Publishing
Plugins for content creation and publishing.
Plugin | Description |
---|---|
Blog | Abstract blog service interface for multiple platforms |
WordPress | WordPress integration via REST API |
Ghost.io | Ghost.io integration via Admin and Content APIs |
NewsRPM | NewsRPM API integration for news article management |
💰 Financial & Data
Plugins for financial data and market information.
Plugin | Description |
---|---|
CloudQuote | Financial data tools for market quotes and news |
🔌 Integration
Protocol and integration plugins.
Plugin | Description |
---|---|
MCP | Model Context Protocol support for standardized AI interactions |
🚀 Getting Started
Installation
Plugins are included in the TokenRing Coder monorepo. Install dependencies:
bun install
Using Plugins
Plugins are automatically loaded based on your configuration. To use a specific plugin:
import { Agent } from '@tokenring-ai/agent';
import WikipediaService from '@tokenring-ai/wikipedia';
const agent = new Agent({
services: [new WikipediaService()],
});
Configuration
Configure plugins in .tokenring/coder-config.mjs
:
export default {
defaults: {
agent: "teamLeader",
model: "gpt-4o"
},
models: {
openai: {
displayName: "OpenAI",
apiKey: process.env.OPENAI_API_KEY
}
}
};
📚 Plugin Architecture
All plugins follow a consistent architecture:
- TokenRingService: Base interface for service plugins
- Tools: Agent-executable functions with Zod validation
- Chat Commands: Interactive commands for agent conversations
- Hooks: Lifecycle hooks for agent events
Common Patterns
- Service Registration: Register services with the agent
- Tool Execution: Execute tools via agent.executeTool()
- Chat Commands: Use /command syntax in conversations
- Event Hooks: Subscribe to agent lifecycle events
🔗 Dependencies
Most plugins depend on:
@tokenring-ai/agent@0.1.0
- Core agent frameworkzod@^4.0.17
- Schema validation
Check individual plugin documentation for specific dependencies.
📖 Next Steps
- Browse individual plugin documentation for detailed usage
- Check out the Quick Start Guide
- Explore example configurations
- Join the community for support and contributions