Rustic AI Agents Documentation¶
Welcome to the Rustic AI Agents documentation. This section provides detailed information about the agents available in the Rustic AI framework.
What is an Agent?¶
In Rustic AI, an agent is a specialized component that performs specific tasks within a guild (multi-agent system). Each agent:
- Processes specific types of messages
- Maintains its own state
- Can interact with external services
- Communicates with other agents through a message-based system
Available Agents¶
Rustic AI includes several specialized agents for various tasks:
Web and API Agents¶
- PlaywrightScraperAgent - Web scraping using Playwright
- SERPAgent - Search engine results via SerpAPI
Language Model Agents¶
- LiteLLMAgent - Interface to various LLMs via LiteLLM
- SimpleLLMAgent - Basic LLM agent implementation
- LLMPhiAgent - Interface to Phi-2 language model
AI Media Generation¶
- RunwaymlStableDiffusionAgent - Image generation using Stable Diffusion
- Image2ImageAgent - Image-to-image transformation with Pix2Pix
- SpeechT5TTSAgent - Text-to-speech synthesis
NLP Agents¶
- SquadAgent - Question answering using the SQuAD model
- MarvinAgent - Classification and data extraction
Utility Agents¶
- VectorAgent - Document indexing and similarity search
- ProbeAgent - Testing utility for monitoring agent interactions
System Agents¶
- UserProxyAgent - Interface between users and the guild
- GuildManagerAgent - Manages guild state and operations
Creating Custom Agents¶
For guidance on creating your own agents, see the Creating Your First Agent guide.