fast-agent
by evalstate
fast-agent enables you to create and interact with sophisticated Agents and Workflows in minutes. It supports complete MCP Feature support including Sampling and both Anthropic and OpenAI models.
Last updated: N/A
What is fast-agent?
fast-agent is a framework for creating and interacting with sophisticated AI Agents and Workflows. It simplifies the development process with a declarative syntax, allowing you to focus on composing prompts and MCP servers.
How to use fast-agent?
To get started, install fast-agent-mcp using uv pip install fast-agent-mcp. Then set up an example agent and config files with fast-agent setup and run your first agent using uv run agent.py. You can easily define agents, combine them into workflows, and use MCP Servers defined in fastagent.config.yaml.
Key features of fast-agent
Simple declarative syntax
Complete MCP Feature support including Sampling
Support for Anthropic and OpenAI models
Multi-modal support (Images and PDFs)
Rapid development and testing with passthrough/playback LLMs
Workflow orchestration (Chain, Parallel, Evaluator-Optimizer, Router, Orchestrator)
Use cases of fast-agent
Building effective agents
Creating researcher agents with evaluator-optimizer workflows
Developing data analysis agents
Automating tasks
Ensembling ideas from different LLMs
FAQ from fast-agent
How do I install fast-agent?
How do I install fast-agent?
Use the command uv pip install fast-agent-mcp.
How can I create an example agent?
How can I create an example agent?
Run fast-agent setup to generate example agent and configuration files.
How do I run an agent?
How do I run an agent?
Use the command uv run agent.py.
How do I specify a model?
How do I specify a model?
Use the --model switch when running an agent, e.g., uv run agent.py --model sonnet.
Where does fast-agent look for configuration files?
Where does fast-agent look for configuration files?
It searches for configuration files in the current directory and then parent directories recursively.