temporal ai agent
Basic Information
This repository is a demo and reference implementation that runs a multi-turn AI agent inside a Temporal workflow. The agent is designed to collect information and execute steps toward a defined goal by running tools and soliciting clarifications as needed. It demonstrates both native tools implemented in the codebase and integration with external services via the Model Context Protocol (MCP). The project supports a default single-agent mode and an experimental multi-agent/multi-goal mode where different agent types can be selected or switched during conversations. Goals are organized in the /goals/ directory by category and the README and docs provide architecture, setup, and customization guidance. The repo shows how to configure LLM providers through LiteLLM-compatible models and includes demo videos and documentation to illustrate how the interaction and workflows operate.
Links
App Details
-
$$$$ -
$$$$