Fast Agent

About

Fast Agent is a Python framework designed to streamline the creation and interaction with sophisticated Agents and Workflows, featuring complete, end-to-end tested MCP support including Sampling, with support for Anthropic and OpenAI models, and multi-modal functionality for handling images and PDFs. Using a simple declarative syntax, it enables developers to focus on composing Prompts and MCP Servers to build effective agents, and offers features like agent chaining, human input requests, parallel processing, and evaluator-optimizer workflows.

Key Features

• Complete MCP feature support, including sampling
• Multi-modal support for images and PDFs
• Declarative syntax for defining agents and workflows
• Evaluator-optimizer workflows
• 278 GitHub stars
• Agent chaining and parallel processing

Use Cases

• Automating complex tasks using workflows
• Rapid prototyping and testing of agent applications
• Building conversational AI agents