Agent Factory is a powerful tool for generating AI agents and workflows using the Model Context Protocol (MCP). It transforms natural language descriptions into executable Python code for agentic workflows, powered by the any-agent library.
Generating your own, complete, executable agentic workflow can be done in just a single command with a single prompt:
uv run agent-factory "Create an AI agent that searches the web for the latest news
on open-source AI and creates a newsletter article summarizing the findings."-
Complete Documentation including installation guide and use-case examples.
-
Natural Language to Code: Transform natural language prompts directly into executable Python code.
-
Multi-Turn and One-Shot Workflows: Support for both multi-turn conversations and one-shot tasks, enabling flexible interaction patterns.
-
Model Provider-Agnostic: By leveraging the any-agent library, it becomes trivial to switch across different agent frameworks and models.
-
Smart Tooling & Integration: Automatically identify the most relevant tools and MCP servers for your agent.
-
Deployable agents: Create runnable agents complete with their operational instructions and dependency files.
-
Agent Evaluation: Evaluate the generated agents against automatically or manually defined criteria to ensure they meet the desired performance standards.
See the LICENSE file for details.